A formative assessment system in Baduanjin physical education based on inertial measurement unit motion capture

A formative assessment system in Baduanjin physical education based on inertial measurement unit motion capture | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A formative assessment system in Baduanjin physical education based on inertial measurement unit motion capture Shanguang Zhao, Yanqing Shen, Ke Zhou, Hai Li This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4323077/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This research aims to use the commercial inertial measurement unit (IMU) motion capture (MoCap) to build a formative assessment system in Baduanjin physical education (PE) to assist PE teachers in completing the formative assessment. The system recognizes motions and assesses motion accuracy by analyzing the motion data captured from students with IMU MoCap. The Baduanjin motions dataset was obtained by recruiting students and teachers to verify the feasibility of the commercial IMU MoCap in recognizing motion accuracy. Furthermore, based on the dataset, suitable methods for recognizing motions and assessing the motion accuracy of Baduanjin were developed and verified. The formative assessment system in Baduanijn PE was built on selected methods and for objective users test. The system obtained excellent accuracy in recognizing student motions (99.77%), and there is a solid connection (over 0.8) between the system and the teacher for assessing motion accuracy. The objective user test demonstrates that the formative assessment system built on IMU MoCap is appropriate for the Baduanjin PE. Health sciences/Health care/Quality of life Physical sciences/Engineering/Biomedical engineering Physical education Baduanijn inertial measurement unit motion capture formative assessment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction For over two decades, the Chinese government has strengthened its support for traditional Chinese sports and included them as recommended sports items in college and university curricula 1 . Baduanjin, a traditional Chinese sport consisting of eight motions, has been circulating for hundreds of years and is highly popular in the recommended sports items 2 . However, universities in China consistently maintain a significant number of students, resulting in a comparatively insufficient number of physical education (PE) teachers, and the number of students enrolled in PE courses surpasses a fair quantity 3 . Due to the many students in PE class, teachers will be unable to provide feedback and correct errors in motions for each student, causing students to repeat the same errors in motions 4 . Besides, formative assessment has been continuously strengthened in the existing physical education assessment system, while the importance of summative assessment, which is the traditional assessment, has been continuously weakened 5 , 6 . However, because of the excessive number of students, PE teachers are unable to conduct a formative assessment, which requires PE teachers to monitor and assess each student's motions. As a result, there is a requirement to assist students in recognising the accuracy and errors in their motions and supporting PE teachers in completing formative assessments for students. Motion capture (MoCap) has been used in sports to monitor athletes' motions in real-time. Some researchers 7 , 8 , for example, compare the motion data collected by MoCap with the data from the standard motions to assess the motion accuracy or detect errors in the motions. Finally, this information is given to athletes and coaches to assist them in correcting errors of motion. Therefore, MoCap was used in the research to build a system for recognising motions and assessing the motion accuracy of Baduanjin, assisting teachers in completing formative assessments. IMU MoCap was chosen to capture motion data. Currently, four different MoCap technologies are typically utilized in sports or PE., including optoelectronic systems (OMS), electromagnetic systems (EMS), image processing systems (IMS), and IMU. However, OMS and IMS have a high cost issue that makes them unsuitable for PE 9 . The shortcoming of EMS is that it is subject to interference from the electromagnetic environment, which is challenging to eliminate in real-world PE applications 10 . The benefit of IMU is that it has a wide measuring range, allows the subject to exercise in a wide area, and is inexpensive 11 . Therefore, this research has two specific objectives. One is to build a formative assessment system in Baduanjin PE for recognising motions and assessing motion accuracy. Another is to use the built formative assessment system in the teaching-learning activities of Baduanjin, and the efficacy of the system has been preliminarily validated. 2. Methods 2.1 Overviews One commercial IMU MoCap: Perception Neuron 2.0 (Noitom Technology, 2018), whose measurement accuracy has been confirmed 12 , was applied in this research. However, the IMU MoCap can effectively distinguish the motion accuracy of Baduanjin motions is unknown. Therefore, this research is divided into three sections (Fig. 1 .). In section one, two groups of students with different motor accuracy levels of Baduanjin were recruited to verify the ability of IMU MoCap to distinguish the difference between the motion accuracy. In section two, methods for assessing and recognising Baduanjin motions were developed and verified using a dataset of Baduanjin motions. Students and teachers from a university in southwestern China were recruited to create the database of Baduanjin motions. In this study, the verified methods are two commonly used types of methods for recognising motions: sample-based and sequence-based methods. The final section applied the selected methods to develop a formative assessment system in Baduanjin PE and evaluated the system's efficacy in Baduanjin courses. Students were recruited to test and record their Baduanjin motions while teaching Baduanjin using the built system. This research was conducted following the Declaration of Helsinki and approved by the Research Ethics Committee of the University of Malaya (UM.TNC2/UMREC-558). Perception Neuron 2.0 contains multiple inertial sensor units, including a three-axis gyroscope, three-axis accelerometer, and three-axis magnetometer 13 . A total of 17 inertial sensing units were used in the investigation. The output file for the motion data captured by Perception Neuron 2.0 is the Biovision Hierarchy (BVH) file generated by the supporting software (Axis Neuron) of Perception Neuron 2.0. The BVH file format was established by the BVH company to store skeleton information and motion data 14 . In this research, the rotation data for each skeleton point in the BVH of motions was extracted to identify and assess motions. BVH skeletons are formed of 17 skeleton points, and rotation data is expressed in Euler angles. The rotation data was transformed into quaternions because Euler angles have gimbal lock and singularity 2 , 11 . 2.2 Effectiveness of Perception Neuron 2.0 in distinguishing motion accuracy of Baduanjin motions This research applied to capture motions with different motion accuracy and compare the differences between the motion data to verify whether Perception Neuron 2.0 can effectively distinguish Baduanjin motions with different motion accuracies. Two groups of undergraduate students in a university in Southwest China with different levels in Baduanjin were recruited. One group was the novice group included nine students who had not taken the Baduanjin course and had no experience in exercising Baduanjin, and the other group was the senior group included 11 students who had taken the Baduanjin course and passed the course test. Another professional teacher with more than ten years of teaching experience was invited as a standard motions model. All participants completed three motion captures using Perception Neuron 2.0. Informed consent was obtained from all subjects . The novice group captured the motions immediately after 30 minutes of initial learning Baduanjin. The captured motion data were converted into quaternions and used dynamic time warping (DTW) to calculate the distances between the standard motions (captured from the invited teacher) and the motions of the two student groups to assess the motion accuracy of the students’ motions 8 . Since the captured motions consist of 17 skeleton points, it is necessary to calculate the distances between the corresponding skeletal points through DTW and then average the distances. Besides, considering the issue of the wrong matching by excessive time warping in DTW, the global warping window was set as 10% of the entire window span in this research to constrain the warp path to be near the diagonal of the matrix 15 . Based on the calculated distances, IBM SPSS Statistics 25.0 as a platform, using independent samples T-test for normally distributed data and Mann-Whitney U test for normally distributed data to evaluate the significance of the differences in motion accuracy between motions (Fig. 2 .) 2.3 Developing and verifying methods for assessing and recognising motions In the research, two commonly different types of methods (the sample-based and the sequence-based methods) were applied in assessing the motion accuracy and recognising the motions of Baduanjin. In the sample-based methods, the features were extracted from the motion data and reduced reduced-dimensionality to prevent data redundancy. In sequence-based methods, keyframes were extracted to prevent data redundancy and reduce processing time. Both methods were classification methods by supervised learning based on the dataset of Baduanjin motions captured by students and teachers. Motions are assigned labels on motion accuracy or motion name. The methods used the motion data to train the classifiers based on labels. After the model parameters of the classifier were trained, unlabeled motions could be classified. Experts of Baduanjin were invited to grade the motion accuracy of the captured motions, and then use the assessment result as labels to train the classifiers. 2.3.1 Developing dataset of Baduanjin motions In this research, the dataset of Baduanjin motions was captured from the undergraduate students, including 20 students and one professional teacher recruited in the first section and 35 students recruited later. The second batch of recruited students captured Baduanjin motions one time. Therefore, the Baduanjin dataset consists of 784 motions, of which 760 motions for students and 24 motions for teachers, and each Baduanjin movement has 98 motions. Two professional Chinese martial art teachers with more than ten years of experience teaching Baduanjin at the university were invited to assess the motion accuracy according to the videos recorded when capturing the students' motions. The assessment applied the grading method used in the Baduanjin course that the motion accuracy of Baduanjin motions is divided into three grades: Fail, Pass, and Good. The Kendall test for the assessment results of the two teachers shows that the Kendall consistency of the evaluation of the eight movements of Baduanjin is above 0.8, indicating that the assessment of the teachers is highly consistent. 2.3.2 The sampled-based methods In the sampled-based methods, each motion is represented by features taken from motion data. The multiple classifier models are trained on the extracted features and the labels of motions and used to classify unlabeled motions. Previous researchers have used time-domain, frequency-domain, and wavelet features to extract features to recognise motions 16 , 17 . This research obtained time-domain features such as mean, variance, standard deviation, skewness, kurtosis, and quartile deviation. Since the motion data comprises 17 skeleton points, the features extracted by one motion data were: 17×3×6 = 306. The extracted features need to be normalized and reduced dimensionality for subsequent training models. The extracted features were normalized to the range [0, 1], and Principal component analysis (PCA) was used for dimensionality reduction of features 18 . Based on the features and labels of motions, the classifiers, including k -Nearest Neighbor ( k -NN), Support Vector Machines (SVM), Naive Bayes (NB), Logistic Regression, Decision Tree (DT), Back Propagation neural network (BPNN), Radial basis function neural network (RBFNN) and One-dimensional CNN (1D-CNN) were trained to assess and recognise motions. The sampled-based methods involved in this research were constructed and verified on Matlab 2020b as the platform. 2.3.3 The sequence-based methods The difference from sample-based methods is that frequency-based methods do not extract features but analyze motion data on quaternions as time-series data. Considering the limited storage space and bandwidth capacity available to users in teaching, extracting keyframes is used to reduce motion data to improve application adaptability. The study used k-means clustering to extract keyframes that can obtain the keyframes corresponding to the compression rate by preset compression rate 19 . In order to confirm the effection of extracting keyframes and obtain a reasonable compression ratio, the interpolation method was applied for reconstructing motions, calculating motion reconstruction error, and then setting the extraction compression ratio to 15% 20, 21 . Based on the keyframes and labels of motions, the sequence-based methods, including DTW combined with classifiers ( k -NN, SVM, NB, Logistic Regression, DT, BPNN, and RBFNN), Hidden Markov Model (HMM), and Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLTSM), and Gated Recurrent Units (GRU) applied to train models to assess and recognise motions. The frequency-based methods involved in this research were constructed and verified on Matlab 2020b as the platform. 2.4 Developing a formative assessment system and taking objective user test The selected methods suitable for assessing the motion accuracy and recognising motions of Baduanjin by verifying the sample-based and sequence-based methods were used to develop a formative assessment system in Baduanjin PE. The formative assessment system included two main functions: one was to assess the motion accuracy of a single motion, and the other was to find the missing motions and sequence errors in motion sequences by recognising motions. The objective user test aims to evaluate the effect of the developed formative system in the Baduanjin PE. Thus, the test was designed to evaluate whether the selected motions in assessing and recognising methods can be used for the formative assessment system in teaching Baduanjin. Undergraduate students with no experience of Baduanjin, without a disability, and no clinical or mental illness were recruited to learn motions according to the Baduanjin PE. After each lesson, the learning progress of students was tested, and the students needed to repeat what they had learned about the motions of Baduanjin. The quality of motions (including the errors in the sequence of motions or missing motions and the motion accuracy) was evaluated using the traditional manual method (by the teacher) and the developed formative assessment system. This process continued throughout the whole cycle of the Baduanjin PE, which took eight weeks to complete the whole teaching and learning process. Finally, the assessment results from the traditional manual method (by teacher) and the developed formative assessment system were compared. 3. Results 3.1 Results of Perception Neuron 2.0 in distinguishing the motion accuracy of Baduanjin motions 63 motion data corresponding to each motion of Baduanjin were obtained by capturing motion three times for 20 students (11 senior students and nine novice students) and one teacher. On the standard motions of the teacher, the distances between the motion data of the students and the teacher at the corresponding 17 skeleton points were calculated using DTW to indicate the accuracy of the student's motion. Independent sample T-tests (for normality data) and Mann-Whitney U tests (for non-normality data) were used to assess novice and senior students' differences in motion accuracy, and the results are shown in Table 1 and Table 2 . Table 1 Differences in motion accuracy between novice and senior students (independent sample T-test) Motion Group N Mean Std. M F Sig. t Sig. a Motion-2 N.S b 27 640.76 74.38 2.289 0.136 4.275 0.000 S.S c 33 565.72 61.64 4.195 0.000 Motion-3 N.S 27 543.46 78.92 4.879 0.031 5.085 0.000 S.S 33 455.75 54.30 4.903 0.000 Motion-4 N.S 27 536.45 41.44 0.061 0.806 5.805 0.000 S.S 33 468.66 47.70 5.888 0.000 a 2-tailed; b novice students; c senior students Table 2 Differences in motion accuracy between novice and senior students (Mann-Whitney U test) Motion Group N Mean Rank Sum of Ranks M-W U a Wilcoxon W Z Asymp. Sig. b Motion-1 N.S c 27 38.52 1040.00 229.00 790.00 -3.217 0.001 S.S d 33 23.94 790.00 Motion-5 N.S c 27 41.96 1133.00 136.00 697.00 -4.599 0.000 S.S d 33 21.12 697.00 Motion-6 N.S c 27 35.93 970.00 299.00 860.00 -2.177 0.029 S.S d 33 26.06 860.00 Motion-7 N.S c 27 37.41 1010.00 259.00 820.00 -2.771 0.000 S.S d 33 24.85 820.00 Motion-8 N.S c 27 42.19 1139.00 130.00 691.00 4.688 0.000 S.S d 33 20.94 691.00 a Mann-Whitney U; b 2-tailed; c novice students; d senior students From Tables 1 and 2 , significant differences ( p < 0.05 or p < 0.01) in motion accuracy of all eight motions between novice and senior students can be found. The differences in motion accuracy between the teacher and senior students were lower than the differences in motion accuracy between the teacher and novice students. The results show that the motion data obtained by Perception Neuron 2.0 can be used to distinguish motions of different motion accuracy of Baduanjin. 3.2 Results of Assessing and recognising motions on the developed methods On the established motion dataset, with the motion scores of two professional Chinese martial art teachers (Teacher A and B) as labels, the accuracy of the two different types of methods under 10-fold cross-validation is shown in Table 4 and Table 5. Table 3 Accuracy of assessing motion accuracy on different methods using the scores of Teacher A as labels Methods Accuracy (%) 1 a 2 3 4 5 6 7 8 Sample-based k -NN 89.47 92.63 b 91.58 b 92.63 b 89.47 b 92.63 b 87.37 88.42 b SVM 89.47 84.21 80.00 92.63 80.00 75.79 95.79 b 80.00 NB 81.05 83.16 74.74 90.53 77.89 80.00 82.11 76.84 Logistic regression 78.95 71.58 62.11 81.05 84.21 77.89 81.05 76.84 DT 73.68 65.26 65.26 61.05 65.26 62.11 73.68 65.26 BPNN 73.68 61.05 63.16 70.53 78.95 66.32 81.05 73.68 RBFNN 83.16 67.37 75.79 75.79 78.95 75.79 84.21 70.53 1D-CNN 71.58 76.84 69.47 76.84 88.42 91.58 78.95 74.74 Sequence-based DTW + k -NN 94.74 b 86.32 77.90 80.00 84.21 77.90 87.37 85.26 DTW + SVM 66.32 62.11 69.47 74.74 63.16 65.26 69.47 78.95 DTW + NB 77.90 72.63 74.74 84.21 65.26 70.53 70.53 74.74 DTW + Logistic regression 66.32 63.16 67.37 73.68 63.16 63.16 66.32 74.74 DTW + DT 69.47 63.16 82.11 70.53 67.37 68.42 74.74 69.47 DTW + BPNN 85.26 71.58 71.58 73.68 66.32 67.37 69.47 84.21 DTW + RBFNN 89.47 84.21 72.63 75.79 80.00 81.05 82.11 83.16 HMM 84.21 80.00 78.95 90.53 76.84 78.95 83.16 77.90 LSTM 75.79 77.90 82.11 84.21 72.63 84.21 78.95 78.95 BiLSTM 84.21 80.00 78.95 90.53 76.84 78.95 83.16 77.90 GRU 80.00 75.79 67.37 83.16 74.74 81.05 82.11 72.63 a Motion; b The highest accuracy Table 4 Accuracy of assessing motion accuracy on different methods using the scores of Teacher B as labels Methods Accuracy (%) 1 a 2 3 4 5 6 7 8 Sample-based k -NN 89.47 86.32 b 88.42 b 91.58 b 91.58 b 86.32 b 85.26 86.32 b SVM 83.16 72.63 74.74 86.32 83.16 84.21 87.37 b 73.68 NB 78.95 78.95 68.42 88.42 80.00 81.05 84.21 75.79 Logistic regression 78.95 71.58 62.11 81.05 84.21 77.89 81.05 76.84 DT 73.68 65.26 65.26 61.05 65.26 62.11 73.68 65.26 BPNN 73.68 61.05 63.16 70.53 78.95 66.32 81.05 73.68 RBFNN 83.16 67.37 75.79 75.79 78.95 75.79 84.21 70.53 1D-CNN 72.63 62.11 76.84 80.00 87.37 81.05 78.95 72.63 Sequence-based DTW + k -NN 92.63 b 77.89 77.90 80.00 83.16 83.16 86.32 83.16 DTW + SVM 66.31 60.00 69.47 69.47 66.32 65.26 72.63 77.90 DTW + NB 75.79 73.68 71.58 71.58 74.74 75.79 74.74 67.37 DTW + Logistic regression 64.21 61.05 61.05 68.42 61.05 62.11 70.53 71.58 DTW + DT 67.37 62.11 60.00 70.53 76.84 76.84 73.68 81.05 DTW + BPNN 68.42 62.11 66.32 65.26 66.32 54.74 71.58 78.95 DTW + RBFNN 78.95 74.74 76.84 62.11 86.32 78.95 86.32 83.16 HMM 83.16 73.68 80.00 80.00 77.90 76.84 82.11 85.26 LSTM 76.84 71.58 77.90 75.79 76.84 82.11 84.21 82.11 BiLSTM 82.11 75.79 78.95 74.74 76.84 82.11 83.16 85.26 GRU 76.84 67.37 69.47 71.58 75.79 78.95 77.90 88.42 a Motion; b The highest accuracy From Table 3 , except the highest accuracy for Motion-1 is DTW + k -NN in the sequence-based method (94.74%) and Motion-7 is the sample-based SVM method (95.79%), for the other six motions, the highest accuracy method is the sample-based k -NN method. From Table 4 , except Motion-1 and Motion-7, the highest accuracy method is the sample-based k -NN method for the other six motions. Although no method can assess all eight motions with the highest accuracy, except for Motion-1 and Motion-7, the accuracy of the sample-based k -NN method is highest for the other six motions. The sample-based k -NN method is the most accurate among all verification methods. Besides, the processing time of the different methods was tested because the methods with excessive processing time are not applied in PE (Table 6 ). Table 6 Processing time of different methods for assessing motion accuracy. Methods Processing time (seconds) Sample-based k -NN 0.008 a SVM 4.751 NB 0.021 Logistics regression 0.020 DT 0.010 BPNN 7.709 RBFNN 0.063 1D-CNN 9.179 Sequence-based DTW + k -NN 3.810 DTW + SVM 4.119 DTW + NB 4.057 DTW + Logistic regression 4.382 DTW + DT 3.947 DTW + BPNN 14.830 DTW + RBFNN 3.898 HMM 4.119 LSTM 14.132 BiLSTM 27.995 GRU 11.943 a Minimum processing time From Table 6 , the processing time of the sample-based k -NN method is the shortest among the validated methods (0.008 seconds). Therefore, considering the accuracy and processing time comprehensively, the method for assessing motion accuracy chose the sample-based k -NN method. For recognising motions, the accuracy and processing time of different methods are shown in Table 7 . Table 7 Accuracy and processing time of different methods for recognising motions. Methods Accuracy (%) Processing time (seconds) Sample-based k -NN 97.63 0.055 b SVM 99.47 0.914 NB 97.89 0.174 Logistics regression 99.21 0.407 DT 84.47 0.087 BPNN 86.97 13.270 RBFNN 75.53 0.295 1D-CNN 99.74 a 80.958 Sequence-based DTW + k -NN 99.47 3.823 DTW + SVM 99.61 6.909 DTW + NB 91.84 6.757 DTW + Logistic regression 94.21 10.163 DTW + DT 93.68 4.809 DTW + BPNN 91.05 24.665 DTW + RBFNN 75.79 5.439 HMM 99.08 61.144 LSTM 96.45 123.477 BiLSTM 97.37 239.190 GRU 97.50 106.513 a The highest accuracy; b Minimum processing time From Table 7 , six methods for recognising motions over 99% accuracy. They were SVM (99.47%), Logistics regression (99.21%) and 1D-CNN (99.74%) in the sample-based methods and DTW + k -NN (99.47%), DTW + SVM (99.61%), and HMM (99.08%) in the sequence-based methods. However, the processing time of the six methods varies widely. The sample-based 1D-CNN method with the highest accuracy is 80.958 seconds, but the sample-based SVM method is 0.914 seconds. The Chi-square test assessed the significant differences in recognising motions of the six methods on the current experimental test results, and the results show there are no significant differences (Table 8 ). Therefore, considering the accuracy and processing time comprehensively, the method for recognising motions chose the sample-based SVM method. Table 8 Chi-square test on the methods for recognising motions over 99% accuracy. Methods Recognised motions Total correct incorrect Sample-based SVM 756 4 760 Logistics regression 754 6 760 1D-CNN 758 2 760 Sequence-based DTW + k-NN 756 4 760 DTW + SVM 757 3 760 HMM 753 7 760 Value of Pearson Chi-Square 4.023 a Asymptotic Significance (2-sided) of Pearson Chi-Square 0.546 a 5 cells (50.0%) have an expected count of less than 6. The minimum expected count is 4.33. 3.3 Results of the objective user test 3.3.1 Developing the formative assessment system Based on the chosen methods, The formative assessment system was developed included two main functions: one function was to assess the motion accuracy of a single motion (using the sample-based k -NN method) and the other function was to confirm whether there were missing motions and sequence errors in motions (using the sample-based SVM method). The steps of the formative assessment system are as follows: Step 1: Input the motion data captured by IMU (BVH file format). Step 2: Extract skeleton data and convert it into quaternion. Step 3: Extract feature values and use PCA to reduce the dimensionality. Step 4: Assess the accuracy of the motions using the trained model on k -NN and recognise the motions using the trained model on SVM. Step 5: Compare the recognised motions with the correct sequence of motions to identify the accuracy of the motion sequence or find the missing motions and sequence errors. Step 6: Output results according to user requirements. 3.3.2 Objective user test The objective user test on the developed formative assessment system has been conducted in a university in Southwest China. Ten students who needed to take the Baduanjin PE as part of their curriculum were recruited, and they used the developed formative assessment system after each lesson. The motions of students were recorded, and the teacher assessed their motions based on this recorded video. The content of the Baduanjin PE is shown in Table 9 . Table 9 The content of the Baduanjin PE. Lesson Content 1 Learn Motion-1 to Motion-3 of Baduanjin 2 Practice Motion-1 to Motion-3 of Baduanjin 3 Practice Motion-1 to Motion-3 of Baduanjin 4 Learn Motion-4 to Motion-6 of Baduanjin 5 Practice Motion-1 to Motion-6 of Baduanjin 6 Learn Motion-7 to Motion-8 of Baduanjin 7 Practice Motion-1 to Motion-8 of Baduanjin 8 Practice Motion-1 to Motion-8 of Baduanjin It can be seen from Table 9 that the Baduanjin PE consists of eight lessons and the entire course lasts for eight weeks (one lesson a week). Learning and practising processes are performed in different weeks, which mean it is not necessary to learn new motions in every lesson. Therefore, a total of 45 motions data were captured for a student, are as shown in Table 10 . Table 10 The captured motions of a student in the objective user test Motion Number of motions In Lesson Motion-1 8 1, 2, 3, 4, 5, 6, 7, 8 Motion-2 8 1, 2, 3, 4, 5, 6, 7, 8 Motion-3 8 1, 2, 3, 4, 5, 6, 7, 8 Motion-4 5 4, 5, 6, 7, 8 Motion-5 5 4, 5, 6, 7, 8 Motion-6 5 4, 5, 6, 7, 8 Motion-7 3 6, 7, 8 Motion-8 3 6, 7, 8 Although a total of 450 motions data (45 data × 10 students = 450 data) could have been obtained to verify the established formative assessment system, only 439 motions data were recorded because some students missed the motions in the learning process. It is found that the developed system wrongly recognised Motion-5 as Motion-3 for one student. In general, the recognition rate was still excellent, reaching 99.77%. Figure 3 . is the output of the recognition results of Motion-3 for all students from the formative assessment system. From the figure, it can be found that there are four missing motions in the recorded data of Motion-3 for three students. Student ID 1 and 4 had one missing motion (Lesson 4), and Student ID 3 had two missing motions (Lessons 4 and 5). All three missing motions happened in Lesson 4 when students learned new motions. The formative assessment system can output the captured motions of any particular student to analyse errors in the sequence of motions and missing motions. Figure 4 . shows the captured motions using the developed formative system to analyze the motions of Student ID 1. It can be seen that Student ID 1 completed all motions assessments for the first three lessons. However, after learning the new motions (Motion-4 to Motion-6) in Lesson 4, the student missed Motion-3 in the assessment. Moreover, after learning the new motion (Motion-7 and Motion-8) in Lesson 6, the student missed Motion-6 in the assessment. The phenomenon of “forgetting motion” reappeared in Lesson 4 and Lesson 6. The effectiveness of assessing the motions of the formative assessment system was tested. The correlation between the assessment results using the developed system versus the assessment results graded by the teacher was obtained, as shown in Table 11 . Table 11 The assessment results and consistency analysis (Kendall test) between the system and teacher Motion System Teacher Kendall value Fail Pass Good Fail Pass Good 1 a 14 58 8 16 58 6 0.904 2 6 52 22 5 58 17 0.865 3 9 28 39 9 35 32 0.855 4 6 22 18 7 25 14 0.867 5 4 26 20 5 29 16 0.862 6 5 31 11 7 32 8 0.835 7 3 22 5 3 24 3 0.850 8 NA b 15 15 NA1 15 15 0.867 a Motion-1; b NA – Not Applicable, which only 2 grades for Motion-8 (Pass & Good) In Table 11 , it can be found that there are no captured motions in Fail for motion-8. The teachers explained that Motion-8 is the most uncomplicated motion in Baduanjin, and it is easy to learn. Students can master the essentials of Motion-8 through short-term learning, but long term practice is needed to achieve high motion accuracy. Therefore, there are no captured motions in Fail. As shown in Table 11 in assessing motion accuracy of Baduanjin, there was high consistency between the formative assessment system and the teacher for all eight motions (all Kendall values were above 0.80). It shows that in the objective user test, the formative assessment system effectively assesses the motion accuracy of students' motions. The formative assessment system can output the motion accuracy of any particular student. Figure 5 . shows the sample result display screen of the motion accuracy results for Student ID 1. From Fig. 5 , it can be found that all the weekly motion accuracy data of Student ID 1 (8 Lessons) has been recorded for the formative assessment. 4. Discussion Formative assessment refers to the assessment to discover the problems of students during the learning process, and it usually consists of a small number of items but requires frequent measurement. Formative assessment can assess how well students are progressing and provide teachers with important information about managing instruction. In contrast, summative assessment does not consider the development of students and problems in the process of learning and feedback from teachers 5 . Therefore, based on the advantages of formative assessment compared with summative assessment, the combined application of the two assessment methods in PE is beneficial to assess students and improve teaching quality effectively. However, due to the high student-teacher ratio in the current education situation in China, PE teachers are unable to observe and evaluate each student's learning during class. Therefore, in most universities, PE teachers only use summative assessments to evaluate students for traditional Chinese sports such as Baduanjin. MoCap has been used in sports to monitor athletes' motions in real-time, and the commercial IMU MoCap has the advantages of being low-cost, relatively simple to use, and adaptable to the environment. However, few studies have applied it in traditional Chinese sports PE, especially for assessing motion accuracy. Therefore, IMU MoCap has the advantage of being applied in traditional Chinese sports PE such as Baduanjin to assist teachers in taking the formative assessment, but the effectiveness of the commercial IMU MoCap for assessing motion accuracy needed to be confirmed. The results of section one show significant differences in the distance between the two groups of students’ motions and the teacher’s motions that verify that the motion data captured by the chosen commercial IMU MoCap: Perception Neuron 2.0 could effectively distinguish Baduanjin motions with different motion accuracy (Tables 1 and 2 ). Based on the verification results of Section One, Section Two developed and selected the appropriate methods for assessing motion accuracy and recognising the motions of Baduanjin. Two different types of methods (sample-based and sequence-based methods) were applied to assess movement accuracy and recognise the motions of Baduanjin. Using the built dataset of Baduanjin motions, the results show that the sample-based k -NN method was selected for assessing motion accuracy for high accuracy and short processing time (Tables 4 , 5, and 6 ). In recognising motions, although there were several methods with accuracy over 99% (Table 7 ), there is no significant difference in the chi-square test between the methods in the current results (Table 8 ). The sample-based SVM method was selected for recognising motions considering the processing time (Table 7 ). The formative assessment system in Baduanjin PE was developed based on the optimal assessing motion accuracy and recognising motions methods selected in the verification methods. Moreover, the objective user test of the system was carried out. The objective user test results show that the accuracy of the formative assessment system in the motion recognition of students reaches 99.77%. The consistency test (Kendall test) of the formative assessment system and teacher on assessing the motion accuracy of students exceeds 0.8. These objective user test results show that the developed formative assessment system effectively assesses motion accuracy and recognises motions. In addition, using the formative assessment system, problems students face in the learning process can be detected immediately. For example, using recognising the motions of students, the system shows the problem of forgetting motions that often occur in learning Baduanjin. The system detected three students who forgot Motion-3 in learning Baduanjin in Lesson 4. Lesson 4 requires students to learn new motions, which leads some students to forget the previously learned motions when learning the new motions. This phenomenon is seen for Student ID 1. The formative assessment system shows that the student was unable to remember Motion-3 (during Lesson 4) and Motion-6 (during Lesson 6). Lastly, the developed formative assessment system could trace the progress of the learning process. As an example of Student ID 1, the recorded result clearly shows the learning progress of all the motion accuracy throughout the eight-week learning process. Therefore, the objective user test results reflect that the first-generation formative assessment system can assess students in the learning process to discover the mistakes made by students. However, there are limitations in the research. In section two, the size of the built dataset may not be large enough for training the model on supervised, which may be the reason for the low accuracy of the neural networks in assessing motions and recognising motions. In section three, the students and teachers have feedback on several shortcomings of Perception Neuron 2.0 used in the objective user test. First, teachers and students highlighted the complexity of wearing the device because Perception Neuron 2.0 uses transmission lines to connect all the sensors to the central unit. Second, Perception Neuron 2.0 restricts the application of persons with different body types because Perception Neuron 2.0 needs to be worn with specially designed clothes. A person over 2m tall and large could not wear the IMU. Third, there have been cases where transmission lines were separated from the connected sensors due to the excessive movement range of the students. Besides, there are limitations of the formative assessment system that need to be improved. In the objective user test, the teacher pointed out that although the system could assess the motion accuracy, it still could not provide feedback on errors in students’ motions in real-time. 5. Conclusion Based on the results, this research confirmed that applying the commercial IMU MoCap such as Perception Neuron 2.0 can distinguish Baduanjin motions with different motion accuracy. Furthermore, with reasonable methods (the sample-based k -NN method and the sample-based SVM method), the formative assessment for assessing and recognising the motions of Baduanjin can be developed, and the effectiveness of the system has been verified in an objective user test. Declarations Author Contributions writing—original draft preparation, S.G. and H.L.; writing—review and editing Y.S. and K.Z. Availability of Data and Materials The datasets used and/or analysed during the current study are available from the corresponding author H.L. on reasonable request. Acknowledgements We would like to acknowledge the participants who took part in this study. Declarations Ethics Approval and Consent to Participate Funding No funding was provided for this study Consent for Publication All participants consented to the publication of the findings. Competing Interests All authors declare that they have no competing interests. References The Central People's Government of the People's Republic of China Healthy China 2030. http://www.gov.cn/zhengce/2016-10/25/content_5124174.htm (accessed 6 February 2022). Li, H.; Selina, K.; Yap, H. J., Differences in Motion Accuracy of Baduanjin between Novice and Senior Students on Inertial Sensor Measurement Systems. Sensors 2020, 20 (21), 6258. Department of Education of Jiangsu Notice of Provincial Education Department on the Results of the Fifth Batch of Public Physical Education Courses in Colleges and Universities https://wenku.baidu.com/view/bb4c0d8f876fb84ae45c3b3567ec102de2bddf83.html (accessed 7 August 2020). Zhan, Y. Y. Exploring a new system of martial arts teaching content in common universities in Shanghai. Master's thesis, East China Normal University, Shanghai, China, 2015. Mastagli, M.; Malini, D.; Hainaut, J. P.; Hoye, A. V.; Bolmont, B., Summative Assessment versus Formative Assessment: An Ecological Study of Physical Education by Analyzing State-Anxiety and Shot-Put Performance among French High School Students. Journal of Physical Education & Sport 2020, 20 (3), 2220–2229, 2020. David, H. A.; Andrés, P. P.; Víctor, L. P., The impact of formative and shared or co-assessment on the acquisition of transversal competences in higher education. Assessment & Evaluation in Higher Education 2019, 44 (6), 933–945. Yamaoka, K.; Uehara, M.; Shima, T.; Tamura, Y., Feedback of flying disc throw with Kinect and its evaluation. Procedia Computer Science 2013, 22 , 912–920. Chen, X. M.; Chen, Z. B.; Li, Y.; He, T. Y.; Hou, J. H.; Liu, S.; He, Y., ImmerTai: immersive motion learning in VR environments. Journal of Visual Communication and Image Representation 2019, 58 , 416–427. van der Kruk, E.; Reijne, M. M., Accuracy of human motion capture systems for sport applications; state-of-the-art review. European Journal of Sport Science 2018, 18 (6), 806–819. Schuler, N.; Bey, M.; Shearn, J.; Butler, D., Evaluation of an electromagnetic position tracking device for measuring in vivo, dynamic joint kinematics. Journal of Biomechanics 2005, 38 (10), 2113–2117. Yap, H. J.; Taha, Z.; Dawal, S. Z. M. A., A generic approach of integrating 3D models into virtual manufacturing. Journal of Zhejiang University-SCIENCE C (Computers & Electronics) 2012, 13 (01), 22–30. Sers, R.; Forrester, S.; Zecca, M.; Ward, S.; Moss, E., Objective assessment of surgeon kinematics during simulated laparoscopic surgery: a preliminary evaluation of the effect of high body mass index models. International Journal of Computer Assisted Radiology and Surgery 2022, 17 , 75–83. Noitom Technology Perception Neuron 2.0. https://www.noitom.com.cn/perception-neuron-2-0.html (accessed 20 August 2020). Dai, H.; Cai, B.; Song, J.; Zhang, D. Y. In Skeletal animation based on BVH motion data , 2nd International Conference on Information Engineering and Computer Science, Wuhan, China, IEEE: Wuhan, China, 2010; pp 1–4. Zhang, W. J.; Wang, J. J.; Zhang, X.; Zhang, K.; Ren, Y. In A novel cardiac arrhythmia detection method relying on improved DTW method , 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China, 2017; IEEE: Chongqing, China, 2017; pp 862–867. Altun, K.; Barshan, B.; Tunel, O., Comparative study on classifying human activities with miniature inertial and magnetic sensors. Pattern Recognition 2010, 43 (10), 3605–3620. Khan, A. M.; Lee, Y. K.; Lee, S. Y.; Kim, T. S., A Triaxial Accelerometer-Based Physical-Activity Recognition via Augmented-Signal Features and a Hierarchical Recognizer. IEEE Transactions on Information Technology in Biomedicine 2010, 14 (5), 1166–1172. Subasi, A.; Gursoy, M. I., EEG signal classification using PCA, ICA, LDA and support vector machines. Expert Systems with Applications 2010, 37 (12), 8659–8666. Shi, X. B.; Liu, S. P.; Zhang, D. Y., Human action recognition method based on key frames. Journal of System Simulation 2015, 27 (10), 2401–2408. Li, H.; Selina, K.; Yap, H. J., Implementation of Sequence-Based Classification Methods for Motion Assessment and Recognition in a Traditional Chinese Sport (Baduanjin). 2022, 19 (3), 1744. Li, S. Y.; Hou, J.; Gan, L. Y., Extraction of motion key-frame based on inter-frame pitch. Computer Engineering 2015, 41 (2), 242–247. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4323077","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":297239518,"identity":"5736721d-51de-4c8a-b56a-ec0e5b7b0a40","order_by":0,"name":"Shanguang Zhao","email":"","orcid":"","institution":"Shanghai Maritime University","correspondingAuthor":false,"prefix":"","firstName":"Shanguang","middleName":"","lastName":"Zhao","suffix":""},{"id":297239524,"identity":"69e2c451-426e-4dc6-a7cc-1f9b85b5b99f","order_by":1,"name":"Yanqing Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABA0lEQVRIie3OsUrEMBjA8YRAugS75jg4X+GDQkAsfZaGQN0O4RY3exRyi+h6gg8huDjm6NClD1BwqQhODr3FSdCkm2BDR4f8CQmB/PiCUCj0D6OENEMOfBUjXLo1xn3kJNIFGi7TZFHOJSvWCrwfCvlo7G0WoTwXbwxq/NQp3ePndA2GHF4YytYecpFYQkQnd4DbYgOGqnOG1MZDzNISKtqD5ljX9odMLBkyspwksnSEJTcj+bYk/vQTVtPFHgoO0dYR46ZQP4k0gQFS4M12B1IreV/T5OwB1CQ5reJjn3/x67sqeu+POpO3TfXafVxlk+TXRJS7g7gNZrwfSSgUCoX+6gcOkVeIOMXU3wAAAABJRU5ErkJggg==","orcid":"","institution":"Henan University","correspondingAuthor":true,"prefix":"","firstName":"Yanqing","middleName":"","lastName":"Shen","suffix":""},{"id":297239529,"identity":"2fbe7625-8c3c-4a25-a058-049203d3c5e2","order_by":2,"name":"Ke Zhou","email":"","orcid":"","institution":"Henan University","correspondingAuthor":false,"prefix":"","firstName":"Ke","middleName":"","lastName":"Zhou","suffix":""},{"id":297239533,"identity":"328d34a2-8add-473d-b261-caa93d304b5c","order_by":3,"name":"Hai Li","email":"","orcid":"","institution":"Neijiang Normal University","correspondingAuthor":false,"prefix":"","firstName":"Hai","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2024-04-25 09:27:04","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4323077/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4323077/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":55787188,"identity":"2f6298dc-3c16-4b84-bc96-8b4363cbfad9","added_by":"auto","created_at":"2024-05-03 07:56:34","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":262539,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of the study\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/69f7a8c9ac0519867b358789.jpg"},{"id":55787190,"identity":"872e435b-3d8c-4f5d-96f3-98ffc070cdc5","added_by":"auto","created_at":"2024-05-03 07:56:34","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":144581,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of verifying the effectiveness of Perception Neuron 2.0 in distinguishing motion accuracy of Baduanjin motions\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/f9e65535d6d83b9b20da9820.jpg"},{"id":55787189,"identity":"37721300-73aa-414a-b097-8caecda1bb75","added_by":"auto","created_at":"2024-05-03 07:56:34","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":38610,"visible":true,"origin":"","legend":"\u003cp\u003eThe formative assessment system displays the student motions for Motion-3\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/152446f056f952a83b15c208.jpg"},{"id":55787187,"identity":"1c7bf72d-bf0f-4713-9634-c9711e9a5bda","added_by":"auto","created_at":"2024-05-03 07:56:34","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":35189,"visible":true,"origin":"","legend":"\u003cp\u003eThe formative assessment system displays the captured motions of Student ID 1\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/5ff5df53855b384077f2100d.jpg"},{"id":55787191,"identity":"edab4a65-1eed-4229-8d47-f8b3e7b0dfaf","added_by":"auto","created_at":"2024-05-03 07:56:34","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":32000,"visible":true,"origin":"","legend":"\u003cp\u003eThe motion accuracy of the formative assessment system for Student ID 1\u003c/p\u003e","description":"","filename":"Picture5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/863d1ba06d85b209e9bd6b7e.jpg"},{"id":60960726,"identity":"5f67e2a4-0cb5-4749-bd2d-2d14d567622e","added_by":"auto","created_at":"2024-07-24 04:50:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1612581,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4323077/v1/00bed672-4f81-4ded-b556-c18baca1d605.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A formative assessment system in Baduanjin physical education based on inertial measurement unit motion capture","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eFor over two decades, the Chinese government has strengthened its support for traditional Chinese sports and included them as recommended sports items in college and university curricula \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Baduanjin, a traditional Chinese sport consisting of eight motions, has been circulating for hundreds of years and is highly popular in the recommended sports items \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. However, universities in China consistently maintain a significant number of students, resulting in a comparatively insufficient number of physical education (PE) teachers, and the number of students enrolled in PE courses surpasses a fair quantity \u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. Due to the many students in PE class, teachers will be unable to provide feedback and correct errors in motions for each student, causing students to repeat the same errors in motions \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Besides, formative assessment has been continuously strengthened in the existing physical education assessment system, while the importance of summative assessment, which is the traditional assessment, has been continuously weakened \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. However, because of the excessive number of students, PE teachers are unable to conduct a formative assessment, which requires PE teachers to monitor and assess each student's motions. As a result, there is a requirement to assist students in recognising the accuracy and errors in their motions and supporting PE teachers in completing formative assessments for students.\u003c/p\u003e \u003cp\u003eMotion capture (MoCap) has been used in sports to monitor athletes' motions in real-time. Some researchers \u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, for example, compare the motion data collected by MoCap with the data from the standard motions to assess the motion accuracy or detect errors in the motions. Finally, this information is given to athletes and coaches to assist them in correcting errors of motion. Therefore, MoCap was used in the research to build a system for recognising motions and assessing the motion accuracy of Baduanjin, assisting teachers in completing formative assessments.\u003c/p\u003e \u003cp\u003eIMU MoCap was chosen to capture motion data. Currently, four different MoCap technologies are typically utilized in sports or PE., including optoelectronic systems (OMS), electromagnetic systems (EMS), image processing systems (IMS), and IMU. However, OMS and IMS have a high cost issue that makes them unsuitable for PE \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. The shortcoming of EMS is that it is subject to interference from the electromagnetic environment, which is challenging to eliminate in real-world PE applications \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. The benefit of IMU is that it has a wide measuring range, allows the subject to exercise in a wide area, and is inexpensive \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Therefore, this research has two specific objectives. One is to build a formative assessment system in Baduanjin PE for recognising motions and assessing motion accuracy. Another is to use the built formative assessment system in the teaching-learning activities of Baduanjin, and the efficacy of the system has been preliminarily validated.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Overviews\u003c/h2\u003e \u003cp\u003eOne commercial IMU MoCap: Perception Neuron 2.0 (Noitom Technology, 2018), whose measurement accuracy has been confirmed \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e, was applied in this research. However, the IMU MoCap can effectively distinguish the motion accuracy of Baduanjin motions is unknown.\u003c/p\u003e \u003cp\u003eTherefore, this research is divided into three sections (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.). In section one, two groups of students with different motor accuracy levels of Baduanjin were recruited to verify the ability of IMU MoCap to distinguish the difference between the motion accuracy. In section two, methods for assessing and recognising Baduanjin motions were developed and verified using a dataset of Baduanjin motions. Students and teachers from a university in southwestern China were recruited to create the database of Baduanjin motions. In this study, the verified methods are two commonly used types of methods for recognising motions: sample-based and sequence-based methods. The final section applied the selected methods to develop a formative assessment system in Baduanjin PE and evaluated the system's efficacy in Baduanjin courses. Students were recruited to test and record their Baduanjin motions while teaching Baduanjin using the built system. This research was conducted following the Declaration of Helsinki and approved by the Research Ethics Committee of the University of Malaya (UM.TNC2/UMREC-558).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePerception Neuron 2.0 contains multiple inertial sensor units, including a three-axis gyroscope, three-axis accelerometer, and three-axis magnetometer \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. A total of 17 inertial sensing units were used in the investigation. The output file for the motion data captured by Perception Neuron 2.0 is the Biovision Hierarchy (BVH) file generated by the supporting software (Axis Neuron) of Perception Neuron 2.0. The BVH file format was established by the BVH company to store skeleton information and motion data \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. In this research, the rotation data for each skeleton point in the BVH of motions was extracted to identify and assess motions. BVH skeletons are formed of 17 skeleton points, and rotation data is expressed in Euler angles. The rotation data was transformed into quaternions because Euler angles have gimbal lock and singularity \u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Effectiveness of Perception Neuron 2.0 in distinguishing motion accuracy of Baduanjin motions\u003c/h2\u003e \u003cp\u003eThis research applied to capture motions with different motion accuracy and compare the differences between the motion data to verify whether Perception Neuron 2.0 can effectively distinguish Baduanjin motions with different motion accuracies. Two groups of undergraduate students in a university in Southwest China with different levels in Baduanjin were recruited. One group was the novice group included nine students who had not taken the Baduanjin course and had no experience in exercising Baduanjin, and the other group was the senior group included 11 students who had taken the Baduanjin course and passed the course test. Another professional teacher with more than ten years of teaching experience was invited as a standard motions model. All participants completed three motion captures using Perception Neuron 2.0. Informed consent was obtained from all subjects .\u003c/p\u003e \u003cp\u003eThe novice group captured the motions immediately after 30 minutes of initial learning Baduanjin. The captured motion data were converted into quaternions and used dynamic time warping (DTW) to calculate the distances between the standard motions (captured from the invited teacher) and the motions of the two student groups to assess the motion accuracy of the students\u0026rsquo; motions \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Since the captured motions consist of 17 skeleton points, it is necessary to calculate the distances between the corresponding skeletal points through DTW and then average the distances. Besides, considering the issue of the wrong matching by excessive time warping in DTW, the global warping window was set as 10% of the entire window span in this research to constrain the warp path to be near the diagonal of the matrix \u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Based on the calculated distances, IBM SPSS Statistics 25.0 as a platform, using independent samples T-test for normally distributed data and Mann-Whitney U test for normally distributed data to evaluate the significance of the differences in motion accuracy between motions (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.)\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Developing and verifying methods for assessing and recognising motions\u003c/h2\u003e \u003cp\u003eIn the research, two commonly different types of methods (the sample-based and the sequence-based methods) were applied in assessing the motion accuracy and recognising the motions of Baduanjin. In the sample-based methods, the features were extracted from the motion data and reduced reduced-dimensionality to prevent data redundancy. In sequence-based methods, keyframes were extracted to prevent data redundancy and reduce processing time. Both methods were classification methods by supervised learning based on the dataset of Baduanjin motions captured by students and teachers. Motions are assigned labels on motion accuracy or motion name. The methods used the motion data to train the classifiers based on labels. After the model parameters of the classifier were trained, unlabeled motions could be classified. Experts of Baduanjin were invited to grade the motion accuracy of the captured motions, and then use the assessment result as labels to train the classifiers.\u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.3.1 Developing dataset of Baduanjin motions\u003c/h2\u003e \u003cp\u003eIn this research, the dataset of Baduanjin motions was captured from the undergraduate students, including 20 students and one professional teacher recruited in the first section and 35 students recruited later. The second batch of recruited students captured Baduanjin motions one time. Therefore, the Baduanjin dataset consists of 784 motions, of which 760 motions for students and 24 motions for teachers, and each Baduanjin movement has 98 motions. Two professional Chinese martial art teachers with more than ten years of experience teaching Baduanjin at the university were invited to assess the motion accuracy according to the videos recorded when capturing the students' motions. The assessment applied the grading method used in the Baduanjin course that the motion accuracy of Baduanjin motions is divided into three grades: Fail, Pass, and Good. The Kendall test for the assessment results of the two teachers shows that the Kendall consistency of the evaluation of the eight movements of Baduanjin is above 0.8, indicating that the assessment of the teachers is highly consistent.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.3.2 The sampled-based methods\u003c/h2\u003e \u003cp\u003eIn the sampled-based methods, each motion is represented by features taken from motion data. The multiple classifier models are trained on the extracted features and the labels of motions and used to classify unlabeled motions. Previous researchers have used time-domain, frequency-domain, and wavelet features to extract features to recognise motions \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. This research obtained time-domain features such as mean, variance, standard deviation, skewness, kurtosis, and quartile deviation. Since the motion data comprises 17 skeleton points, the features extracted by one motion data were: 17\u0026times;3\u0026times;6\u0026thinsp;=\u0026thinsp;306. The extracted features need to be normalized and reduced dimensionality for subsequent training models. The extracted features were normalized to the range [0, 1], and Principal component analysis (PCA) was used for dimensionality reduction of features \u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Based on the features and labels of motions, the classifiers, including \u003cem\u003ek\u003c/em\u003e-Nearest Neighbor (\u003cem\u003ek\u003c/em\u003e-NN), Support Vector Machines (SVM), Naive Bayes (NB), Logistic Regression, Decision Tree (DT), Back Propagation neural network (BPNN), Radial basis function neural network (RBFNN) and One-dimensional CNN (1D-CNN) were trained to assess and recognise motions. The sampled-based methods involved in this research were constructed and verified on Matlab 2020b as the platform.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.3.3 The sequence-based methods\u003c/h2\u003e \u003cp\u003eThe difference from sample-based methods is that frequency-based methods do not extract features but analyze motion data on quaternions as time-series data. Considering the limited storage space and bandwidth capacity available to users in teaching, extracting keyframes is used to reduce motion data to improve application adaptability. The study used k-means clustering to extract keyframes that can obtain the keyframes corresponding to the compression rate by preset compression rate \u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. In order to confirm the effection of extracting keyframes and obtain a reasonable compression ratio, the interpolation method was applied for reconstructing motions, calculating motion reconstruction error, and then setting the extraction compression ratio to 15% \u003csup\u003e20, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Based on the keyframes and labels of motions, the sequence-based methods, including DTW combined with classifiers (\u003cem\u003ek\u003c/em\u003e-NN, SVM, NB, Logistic Regression, DT, BPNN, and RBFNN), Hidden Markov Model (HMM), and Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLTSM), and Gated Recurrent Units (GRU) applied to train models to assess and recognise motions. The frequency-based methods involved in this research were constructed and verified on Matlab 2020b as the platform.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Developing a formative assessment system and taking objective user test\u003c/h2\u003e \u003cp\u003eThe selected methods suitable for assessing the motion accuracy and recognising motions of Baduanjin by verifying the sample-based and sequence-based methods were used to develop a formative assessment system in Baduanjin PE. The formative assessment system included two main functions: one was to assess the motion accuracy of a single motion, and the other was to find the missing motions and sequence errors in motion sequences by recognising motions.\u003c/p\u003e \u003cp\u003eThe objective user test aims to evaluate the effect of the developed formative system in the Baduanjin PE. Thus, the test was designed to evaluate whether the selected motions in assessing and recognising methods can be used for the formative assessment system in teaching Baduanjin. Undergraduate students with no experience of Baduanjin, without a disability, and no clinical or mental illness were recruited to learn motions according to the Baduanjin PE. After each lesson, the learning progress of students was tested, and the students needed to repeat what they had learned about the motions of Baduanjin. The quality of motions (including the errors in the sequence of motions or missing motions and the motion accuracy) was evaluated using the traditional manual method (by the teacher) and the developed formative assessment system. This process continued throughout the whole cycle of the Baduanjin PE, which took eight weeks to complete the whole teaching and learning process. Finally, the assessment results from the traditional manual method (by teacher) and the developed formative assessment system were compared.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Results of Perception Neuron 2.0 in distinguishing the motion accuracy of Baduanjin motions\u003c/h2\u003e \u003cp\u003e63 motion data corresponding to each motion of Baduanjin were obtained by capturing motion three times for 20 students (11 senior students and nine novice students) and one teacher. On the standard motions of the teacher, the distances between the motion data of the students and the teacher at the corresponding 17 skeleton points were calculated using DTW to indicate the accuracy of the student's motion. Independent sample T-tests (for normality data) and Mann-Whitney U tests (for non-normality data) were used to assess novice and senior students' differences in motion accuracy, and the results are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDifferences in motion accuracy between novice and senior students (independent sample T-test)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eStd. M\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSig.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003et\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eSig. \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e640.76\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e74.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e2.289\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.136\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e4.275\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e565.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e61.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e4.195\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e543.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e78.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4.879\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e5.085\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e455.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e54.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e4.903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e536.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e41.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.061\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e5.805\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e468.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e47.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e5.888\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"9\"\u003e\u003csup\u003ea\u003c/sup\u003e 2-tailed; \u003csup\u003eb\u003c/sup\u003e novice students; \u003csup\u003ec\u003c/sup\u003e senior students\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDifferences in motion accuracy between novice and senior students (Mann-Whitney U test)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGroup\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMean Rank\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSum of Ranks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eM-W U \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eWilcoxon W\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eZ\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eAsymp. Sig. \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e38.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1040.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e229.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e790.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-3.217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e23.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e790.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e41.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1133.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e136.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e697.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-4.599\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e21.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e697.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e35.93\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e970.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e299.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e860.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-2.177\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e860.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e37.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1010.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e259.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e820.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-2.771\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e24.85\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e820.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion-8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN.S \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e42.19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1139.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e130.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e691.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e4.688\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS.S \u003csup\u003ed\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e20.94\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e691.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"9\"\u003e\u003csup\u003ea\u003c/sup\u003e Mann-Whitney U; \u003csup\u003eb\u003c/sup\u003e 2-tailed; \u003csup\u003ec\u003c/sup\u003e novice students; \u003csup\u003ed\u003c/sup\u003e senior students\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, significant differences (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 or \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01) in motion accuracy of all eight motions between novice and senior students can be found. The differences in motion accuracy between the teacher and senior students were lower than the differences in motion accuracy between the teacher and novice students. The results show that the motion data obtained by Perception Neuron 2.0 can be used to distinguish motions of different motion accuracy of Baduanjin.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Results of Assessing and recognising motions on the developed methods\u003c/h2\u003e \u003cp\u003eOn the established motion dataset, with the motion scores of two professional Chinese martial art teachers (Teacher A and B) as labels, the accuracy of the two different types of methods under 10-fold cross-validation is shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and Table\u0026nbsp;5.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAccuracy of assessing motion accuracy on different methods using the scores of Teacher A as labels\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"8\" nameend=\"c10\" namest=\"c3\"\u003e \u003cp\u003eAccuracy (%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eSample-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e92.63 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e91.58 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e92.63 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e89.47 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e92.63 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e87.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e88.42 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e92.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e95.79 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e90.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e77.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLogistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e77.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1D-CNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e88.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e91.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003eSequence-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94.74 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e86.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e87.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e85.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;NB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW +\u003c/p\u003e \u003cp\u003eLogistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;DT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e68.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;BPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e85.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;RBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHMM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e90.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBiLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e90.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGRU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"10\"\u003e\u003csup\u003ea\u003c/sup\u003e Motion; \u003csup\u003eb\u003c/sup\u003e The highest accuracy\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAccuracy of assessing motion accuracy on different methods using the scores of Teacher B as labels\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"8\" nameend=\"c10\" namest=\"c3\"\u003e \u003cp\u003eAccuracy (%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eSample-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e89.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e86.32 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e88.42 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e91.58 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e91.58 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e86.32 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e85.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e86.32 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e86.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e87.37 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e68.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e88.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLogistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e77.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e63.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1D-CNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e87.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003eSequence-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92.63 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e77.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e86.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e66.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e60.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e72.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;NB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW +\u003c/p\u003e \u003cp\u003eLogistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e64.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e68.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e61.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;DT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e60.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e70.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e81.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;BPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e68.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e65.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e66.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e54.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;RBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e62.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e86.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e86.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHMM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e73.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e80.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e85.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e84.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBiLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e74.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e82.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e83.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e85.26\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGRU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e76.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e67.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e69.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e71.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e78.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003e77.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c10\"\u003e \u003cp\u003e88.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"10\"\u003e\u003csup\u003ea\u003c/sup\u003e Motion; \u003csup\u003eb\u003c/sup\u003e The highest accuracy\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, except the highest accuracy for Motion-1 is DTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN in the sequence-based method (94.74%) and Motion-7 is the sample-based SVM method (95.79%), for the other six motions, the highest accuracy method is the sample-based \u003cem\u003ek\u003c/em\u003e-NN method. From Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, except Motion-1 and Motion-7, the highest accuracy method is the sample-based \u003cem\u003ek\u003c/em\u003e-NN method for the other six motions. Although no method can assess all eight motions with the highest accuracy, except for Motion-1 and Motion-7, the accuracy of the sample-based \u003cem\u003ek\u003c/em\u003e-NN method is highest for the other six motions. The sample-based \u003cem\u003ek\u003c/em\u003e-NN method is the most accurate among all verification methods.\u003c/p\u003e \u003cp\u003eBesides, the processing time of the different methods was tested because the methods with excessive processing time are not applied in PE (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eProcessing time of different methods for assessing motion accuracy.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eProcessing time (seconds)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eSample-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.008 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.751\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLogistics regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7.709\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.063\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1D-CNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.179\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003eSequence-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.810\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.119\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;NB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.057\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;Logistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.382\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;DT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.947\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;BPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.830\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;RBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.898\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHMM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.119\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.132\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBiLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27.995\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGRU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.943\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"3\"\u003e\u003csup\u003ea\u003c/sup\u003e Minimum processing time\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e6\u003c/span\u003e, the processing time of the sample-based \u003cem\u003ek\u003c/em\u003e-NN method is the shortest among the validated methods (0.008 seconds). Therefore, considering the accuracy and processing time comprehensively, the method for assessing motion accuracy chose the sample-based \u003cem\u003ek\u003c/em\u003e-NN method.\u003c/p\u003e \u003cp\u003eFor recognising motions, the accuracy and processing time of different methods are shown in Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e7\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAccuracy and processing time of different methods for recognising motions.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eProcessing time (seconds)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eSample-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e97.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.055 \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.914\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e97.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.174\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLogistics regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.407\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e84.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.087\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e86.97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13.270\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.295\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1D-CNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.74 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.958\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"10\" rowspan=\"11\"\u003e \u003cp\u003eSequence-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.823\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.909\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;NB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e91.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.757\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;Logistic regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e94.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.163\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;DT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e93.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4.809\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;BPNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e91.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24.665\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;RBFNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5.439\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHMM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e61.144\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e96.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e123.477\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBiLSTM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e97.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e239.190\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGRU\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e97.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e106.513\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003e\u003csup\u003ea\u003c/sup\u003e The highest accuracy; \u003csup\u003eb\u003c/sup\u003e Minimum processing time\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e7\u003c/span\u003e, six methods for recognising motions over 99% accuracy. They were SVM (99.47%), Logistics regression (99.21%) and 1D-CNN (99.74%) in the sample-based methods and DTW\u0026thinsp;+\u0026thinsp;\u003cem\u003ek\u003c/em\u003e-NN (99.47%), DTW\u0026thinsp;+\u0026thinsp;SVM (99.61%), and HMM (99.08%) in the sequence-based methods. However, the processing time of the six methods varies widely. The sample-based 1D-CNN method with the highest accuracy is 80.958 seconds, but the sample-based SVM method is 0.914 seconds. The Chi-square test assessed the significant differences in recognising motions of the six methods on the current experimental test results, and the results show there are no significant differences (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Therefore, considering the accuracy and processing time comprehensively, the method for recognising motions chose the sample-based SVM method.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eChi-square test on the methods for recognising motions over 99% accuracy.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eMethods\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eRecognised motions\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ecorrect\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eincorrect\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eSample-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e756\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLogistics regression\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1D-CNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eSequence-based\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;k-NN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e756\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDTW\u0026thinsp;+\u0026thinsp;SVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e757\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHMM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e753\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e760\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eValue of Pearson Chi-Square\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.023 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eAsymptotic Significance (2-sided) of Pearson Chi-Square\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.546\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003csup\u003ea\u003c/sup\u003e 5 cells (50.0%) have an expected count of less than 6. The minimum expected count is 4.33.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Results of the objective user test\u003c/h2\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 Developing the formative assessment system\u003c/h2\u003e \u003cp\u003eBased on the chosen methods, The formative assessment system was developed included two main functions: one function was to assess the motion accuracy of a single motion (using the sample-based \u003cem\u003ek\u003c/em\u003e-NN method) and the other function was to confirm whether there were missing motions and sequence errors in motions (using the sample-based SVM method).\u003c/p\u003e \u003cp\u003eThe steps of the formative assessment system are as follows:\u003c/p\u003e \u003cp\u003eStep 1: Input the motion data captured by IMU (BVH file format).\u003c/p\u003e \u003cp\u003eStep 2: Extract skeleton data and convert it into quaternion.\u003c/p\u003e \u003cp\u003eStep 3: Extract feature values and use PCA to reduce the dimensionality.\u003c/p\u003e \u003cp\u003eStep 4: Assess the accuracy of the motions using the trained model on \u003cem\u003ek\u003c/em\u003e-NN and recognise the motions using the trained model on SVM.\u003c/p\u003e \u003cp\u003eStep 5: Compare the recognised motions with the correct sequence of motions to identify the accuracy of the motion sequence or find the missing motions and sequence errors.\u003c/p\u003e \u003cp\u003eStep 6: Output results according to user requirements.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Objective user test\u003c/h2\u003e \u003cp\u003eThe objective user test on the developed formative assessment system has been conducted in a university in Southwest China. Ten students who needed to take the Baduanjin PE as part of their curriculum were recruited, and they used the developed formative assessment system after each lesson. The motions of students were recorded, and the teacher assessed their motions based on this recorded video. The content of the Baduanjin PE is shown in Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e9\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab8\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 9\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe content of the Baduanjin PE.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLesson\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eContent\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLearn Motion-1 to Motion-3 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePractice Motion-1 to Motion-3 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePractice Motion-1 to Motion-3 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLearn Motion-4 to Motion-6 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePractice Motion-1 to Motion-6 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLearn Motion-7 to Motion-8 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePractice Motion-1 to Motion-8 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePractice Motion-1 to Motion-8 of Baduanjin\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIt can be seen from Table\u0026nbsp;\u003cspan refid=\"Tab8\" class=\"InternalRef\"\u003e9\u003c/span\u003e that the Baduanjin PE consists of eight lessons and the entire course lasts for eight weeks (one lesson a week). Learning and practising processes are performed in different weeks, which mean it is not necessary to learn new motions in every lesson. Therefore, a total of 45 motions data were captured for a student, are as shown in Table\u0026nbsp;\u003cspan refid=\"Tab9\" class=\"InternalRef\"\u003e10\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab9\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 10\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe captured motions of a student in the objective user test\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of motions\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIn Lesson\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1, 2, 3, 4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1, 2, 3, 4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1, 2, 3, 4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4, 5, 6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMotion-8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6, 7, 8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAlthough a total of 450 motions data (45 data \u0026times; 10 students\u0026thinsp;=\u0026thinsp;450 data) could have been obtained to verify the established formative assessment system, only 439 motions data were recorded because some students missed the motions in the learning process. It is found that the developed system wrongly recognised Motion-5 as Motion-3 for one student. In general, the recognition rate was still excellent, reaching 99.77%.\u003c/p\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. is the output of the recognition results of Motion-3 for all students from the formative assessment system. From the figure, it can be found that there are four missing motions in the recorded data of Motion-3 for three students. Student ID 1 and 4 had one missing motion (Lesson 4), and Student ID 3 had two missing motions (Lessons 4 and 5). All three missing motions happened in Lesson 4 when students learned new motions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe formative assessment system can output the captured motions of any particular student to analyse errors in the sequence of motions and missing motions. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. shows the captured motions using the developed formative system to analyze the motions of Student ID 1. It can be seen that Student ID 1 completed all motions assessments for the first three lessons. However, after learning the new motions (Motion-4 to Motion-6) in Lesson 4, the student missed Motion-3 in the assessment. Moreover, after learning the new motion (Motion-7 and Motion-8) in Lesson 6, the student missed Motion-6 in the assessment. The phenomenon of \u0026ldquo;forgetting motion\u0026rdquo; reappeared in Lesson 4 and Lesson 6.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe effectiveness of assessing the motions of the formative assessment system was tested. The correlation between the assessment results using the developed system versus the assessment results graded by the teacher was obtained, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e11\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab10\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 11\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe assessment results and consistency analysis (Kendall test) between the system and teacher\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eMotion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eSystem\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eTeacher\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eKendall value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFail\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePass\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGood\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFail\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePass\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eGood\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1 \u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.904\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.865\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.855\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.867\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.862\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.835\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.850\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNA \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNA1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.867\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003ea\u003c/sup\u003e Motion-1; \u003csup\u003eb\u003c/sup\u003e NA \u0026ndash; Not Applicable, which only 2 grades for Motion-8 (Pass \u0026amp; Good)\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e11\u003c/span\u003e, it can be found that there are no captured motions in Fail for motion-8. The teachers explained that Motion-8 is the most uncomplicated motion in Baduanjin, and it is easy to learn. Students can master the essentials of Motion-8 through short-term learning, but long term practice is needed to achieve high motion accuracy. Therefore, there are no captured motions in Fail. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab10\" class=\"InternalRef\"\u003e11\u003c/span\u003e in assessing motion accuracy of Baduanjin, there was high consistency between the formative assessment system and the teacher for all eight motions (all Kendall values were above 0.80). It shows that in the objective user test, the formative assessment system effectively assesses the motion accuracy of students' motions.\u003c/p\u003e \u003cp\u003eThe formative assessment system can output the motion accuracy of any particular student. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e. shows the sample result display screen of the motion accuracy results for Student ID 1. From Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, it can be found that all the weekly motion accuracy data of Student ID 1 (8 Lessons) has been recorded for the formative assessment.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eFormative assessment refers to the assessment to discover the problems of students during the learning process, and it usually consists of a small number of items but requires frequent measurement. Formative assessment can assess how well students are progressing and provide teachers with important information about managing instruction. In contrast, summative assessment does not consider the development of students and problems in the process of learning and feedback from teachers \u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Therefore, based on the advantages of formative assessment compared with summative assessment, the combined application of the two assessment methods in PE is beneficial to assess students and improve teaching quality effectively. However, due to the high student-teacher ratio in the current education situation in China, PE teachers are unable to observe and evaluate each student's learning during class. Therefore, in most universities, PE teachers only use summative assessments to evaluate students for traditional Chinese sports such as Baduanjin. MoCap has been used in sports to monitor athletes' motions in real-time, and the commercial IMU MoCap has the advantages of being low-cost, relatively simple to use, and adaptable to the environment. However, few studies have applied it in traditional Chinese sports PE, especially for assessing motion accuracy. Therefore, IMU MoCap has the advantage of being applied in traditional Chinese sports PE such as Baduanjin to assist teachers in taking the formative assessment, but the effectiveness of the commercial IMU MoCap for assessing motion accuracy needed to be confirmed. The results of section one show significant differences in the distance between the two groups of students\u0026rsquo; motions and the teacher\u0026rsquo;s motions that verify that the motion data captured by the chosen commercial IMU MoCap: Perception Neuron 2.0 could effectively distinguish Baduanjin motions with different motion accuracy (Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Based on the verification results of Section One, Section Two developed and selected the appropriate methods for assessing motion accuracy and recognising the motions of Baduanjin. Two different types of methods (sample-based and sequence-based methods) were applied to assess movement accuracy and recognise the motions of Baduanjin. Using the built dataset of Baduanjin motions, the results show that the sample-based \u003cem\u003ek\u003c/em\u003e-NN method was selected for assessing motion accuracy for high accuracy and short processing time (Tables\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, 5, and \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e6\u003c/span\u003e). In recognising motions, although there were several methods with accuracy over 99% (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e7\u003c/span\u003e), there is no significant difference in the chi-square test between the methods in the current results (Table\u0026nbsp;\u003cspan refid=\"Tab7\" class=\"InternalRef\"\u003e8\u003c/span\u003e). The sample-based SVM method was selected for recognising motions considering the processing time (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe formative assessment system in Baduanjin PE was developed based on the optimal assessing motion accuracy and recognising motions methods selected in the verification methods. Moreover, the objective user test of the system was carried out. The objective user test results show that the accuracy of the formative assessment system in the motion recognition of students reaches 99.77%. The consistency test (Kendall test) of the formative assessment system and teacher on assessing the motion accuracy of students exceeds 0.8. These objective user test results show that the developed formative assessment system effectively assesses motion accuracy and recognises motions. In addition, using the formative assessment system, problems students face in the learning process can be detected immediately. For example, using recognising the motions of students, the system shows the problem of forgetting motions that often occur in learning Baduanjin. The system detected three students who forgot Motion-3 in learning Baduanjin in Lesson 4. Lesson 4 requires students to learn new motions, which leads some students to forget the previously learned motions when learning the new motions. This phenomenon is seen for Student ID 1. The formative assessment system shows that the student was unable to remember Motion-3 (during Lesson 4) and Motion-6 (during Lesson 6). Lastly, the developed formative assessment system could trace the progress of the learning process. As an example of Student ID 1, the recorded result clearly shows the learning progress of all the motion accuracy throughout the eight-week learning process. Therefore, the objective user test results reflect that the first-generation formative assessment system can assess students in the learning process to discover the mistakes made by students.\u003c/p\u003e \u003cp\u003eHowever, there are limitations in the research. In section two, the size of the built dataset may not be large enough for training the model on supervised, which may be the reason for the low accuracy of the neural networks in assessing motions and recognising motions. In section three, the students and teachers have feedback on several shortcomings of Perception Neuron 2.0 used in the objective user test. First, teachers and students highlighted the complexity of wearing the device because Perception Neuron 2.0 uses transmission lines to connect all the sensors to the central unit. Second, Perception Neuron 2.0 restricts the application of persons with different body types because Perception Neuron 2.0 needs to be worn with specially designed clothes. A person over 2m tall and large could not wear the IMU. Third, there have been cases where transmission lines were separated from the connected sensors due to the excessive movement range of the students. Besides, there are limitations of the formative assessment system that need to be improved. In the objective user test, the teacher pointed out that although the system could assess the motion accuracy, it still could not provide feedback on errors in students\u0026rsquo; motions in real-time.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eBased on the results, this research confirmed that applying the commercial IMU MoCap such as Perception Neuron 2.0 can distinguish Baduanjin motions with different motion accuracy. Furthermore, with reasonable methods (the sample-based \u003cem\u003ek\u003c/em\u003e-NN method and the sample-based SVM method), the formative assessment for assessing and recognising the motions of Baduanjin can be developed, and the effectiveness of the system has been verified in an objective user test.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ewriting\u0026mdash;original draft preparation, S.G.\u0026nbsp;and H.L.; writing\u0026mdash;review and editing\u0026nbsp;Y.S. and\u0026nbsp;K.Z.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author\u0026nbsp;\u0026nbsp;H.L.\u0026nbsp;on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to acknowledge the participants who took part in this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthics Approval and Consent to Participate\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo funding was provided for this study\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll participants consented to the publication of the findings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eThe Central People's Government of the People's Republic of China Healthy China 2030. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.gov.cn/zhengce/2016-10/25/content_5124174.htm\u003c/span\u003e\u003cspan address=\"http://www.gov.cn/zhengce/2016-10/25/content_5124174.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed 6 February 2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, H.; Selina, K.; Yap, H. J., Differences in Motion Accuracy of Baduanjin between Novice and Senior Students on Inertial Sensor Measurement Systems. Sensors 2020, \u003cem\u003e20\u003c/em\u003e (21), 6258.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDepartment of Education of Jiangsu Notice of Provincial Education Department on the Results of the Fifth Batch of Public Physical Education Courses in Colleges and Universities \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://wenku.baidu.com/view/bb4c0d8f876fb84ae45c3b3567ec102de2bddf83.html\u003c/span\u003e\u003cspan address=\"https://wenku.baidu.com/view/bb4c0d8f876fb84ae45c3b3567ec102de2bddf83.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed 7 August 2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhan, Y. Y. Exploring a new system of martial arts teaching content in common universities in Shanghai. Master's thesis, East China Normal University, Shanghai, China, 2015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMastagli, M.; Malini, D.; Hainaut, J. P.; Hoye, A. V.; Bolmont, B., Summative Assessment versus Formative Assessment: An Ecological Study of Physical Education by Analyzing State-Anxiety and Shot-Put Performance among French High School Students. \u003cem\u003eJournal of Physical Education \u0026amp; Sport\u003c/em\u003e 2020, \u003cem\u003e20\u003c/em\u003e (3), 2220\u0026ndash;2229, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavid, H. A.; Andr\u0026eacute;s, P. P.; V\u0026iacute;ctor, L. P., The impact of formative and shared or co-assessment on the acquisition of transversal competences in higher education. Assessment \u0026amp; Evaluation in Higher Education 2019, \u003cem\u003e44\u003c/em\u003e (6), 933\u0026ndash;945.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYamaoka, K.; Uehara, M.; Shima, T.; Tamura, Y., Feedback of flying disc throw with Kinect and its evaluation. Procedia Computer Science 2013, \u003cem\u003e22\u003c/em\u003e, 912\u0026ndash;920.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, X. M.; Chen, Z. B.; Li, Y.; He, T. Y.; Hou, J. H.; Liu, S.; He, Y., ImmerTai: immersive motion learning in VR environments. Journal of Visual Communication and Image Representation 2019, \u003cem\u003e58\u003c/em\u003e, 416\u0026ndash;427.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Kruk, E.; Reijne, M. M., Accuracy of human motion capture systems for sport applications; state-of-the-art review. European Journal of Sport Science 2018, \u003cem\u003e18\u003c/em\u003e (6), 806\u0026ndash;819.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchuler, N.; Bey, M.; Shearn, J.; Butler, D., Evaluation of an electromagnetic position tracking device for measuring in vivo, dynamic joint kinematics. Journal of Biomechanics 2005, \u003cem\u003e38\u003c/em\u003e (10), 2113\u0026ndash;2117.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYap, H. J.; Taha, Z.; Dawal, S. Z. M. A., A generic approach of integrating 3D models into virtual manufacturing. Journal of Zhejiang University-SCIENCE C (Computers \u0026amp; Electronics) 2012, \u003cem\u003e13\u003c/em\u003e (01), 22\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSers, R.; Forrester, S.; Zecca, M.; Ward, S.; Moss, E., Objective assessment of surgeon kinematics during simulated laparoscopic surgery: a preliminary evaluation of the effect of high body mass index models. International Journal of Computer Assisted Radiology and Surgery 2022, \u003cem\u003e17\u003c/em\u003e, 75\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNoitom Technology Perception Neuron 2.0. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.noitom.com.cn/perception-neuron-2-0.html\u003c/span\u003e\u003cspan address=\"https://www.noitom.com.cn/perception-neuron-2-0.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed 20 August 2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDai, H.; Cai, B.; Song, J.; Zhang, D. Y. In \u003cem\u003eSkeletal animation based on BVH motion data\u003c/em\u003e, 2nd International Conference on Information Engineering and Computer Science, Wuhan, China, IEEE: Wuhan, China, 2010; pp 1\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, W. J.; Wang, J. J.; Zhang, X.; Zhang, K.; Ren, Y. In \u003cem\u003eA novel cardiac arrhythmia detection method relying on improved DTW method\u003c/em\u003e, 2017 IEEE 2nd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China, 2017; IEEE: Chongqing, China, 2017; pp 862\u0026ndash;867.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAltun, K.; Barshan, B.; Tunel, O., Comparative study on classifying human activities with miniature inertial and magnetic sensors. Pattern Recognition 2010, \u003cem\u003e43\u003c/em\u003e (10), 3605\u0026ndash;3620.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhan, A. M.; Lee, Y. K.; Lee, S. Y.; Kim, T. S., A Triaxial Accelerometer-Based Physical-Activity Recognition via Augmented-Signal Features and a Hierarchical Recognizer. IEEE Transactions on Information Technology in Biomedicine 2010, \u003cem\u003e14\u003c/em\u003e (5), 1166\u0026ndash;1172.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubasi, A.; Gursoy, M. I., EEG signal classification using PCA, ICA, LDA and support vector machines. Expert Systems with Applications 2010, \u003cem\u003e37\u003c/em\u003e (12), 8659\u0026ndash;8666.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi, X. B.; Liu, S. P.; Zhang, D. Y., Human action recognition method based on key frames. Journal of System Simulation 2015, \u003cem\u003e27\u003c/em\u003e (10), 2401\u0026ndash;2408.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, H.; Selina, K.; Yap, H. J., Implementation of Sequence-Based Classification Methods for Motion Assessment and Recognition in a Traditional Chinese Sport (Baduanjin). 2022, \u003cem\u003e19\u003c/em\u003e (3), 1744.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, S. Y.; Hou, J.; Gan, L. Y., Extraction of motion key-frame based on inter-frame pitch. Computer Engineering 2015, \u003cem\u003e41\u003c/em\u003e (2), 242\u0026ndash;247.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Physical education, Baduanijn, inertial measurement unit, motion capture, formative assessment","lastPublishedDoi":"10.21203/rs.3.rs-4323077/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4323077/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis research aims to use the commercial inertial measurement unit (IMU) motion capture (MoCap) to build a formative assessment system in Baduanjin physical education (PE) to assist PE teachers in completing the formative assessment. The system recognizes motions and assesses motion accuracy by analyzing the motion data captured from students with IMU MoCap. The Baduanjin motions dataset was obtained by recruiting students and teachers to verify the feasibility of the commercial IMU MoCap in recognizing motion accuracy. Furthermore, based on the dataset, suitable methods for recognizing motions and assessing the motion accuracy of Baduanjin were developed and verified. The formative assessment system in Baduanijn PE was built on selected methods and for objective users test. The system obtained excellent accuracy in recognizing student motions (99.77%), and there is a solid connection (over 0.8) between the system and the teacher for assessing motion accuracy. The objective user test demonstrates that the formative assessment system built on IMU MoCap is appropriate for the Baduanjin PE.\u003c/p\u003e","manuscriptTitle":"A formative assessment system in Baduanjin physical education based on inertial measurement unit motion capture","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-03 07:56:30","doi":"10.21203/rs.3.rs-4323077/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4d214085-5e33-4b4f-ae6c-95f883ad5057","owner":[],"postedDate":"May 3rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":31355717,"name":"Health sciences/Health care/Quality of life"},{"id":31355718,"name":"Physical sciences/Engineering/Biomedical engineering"}],"tags":[],"updatedAt":"2024-07-24T04:42:01+00:00","versionOfRecord":[],"versionCreatedAt":"2024-05-03 07:56:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4323077","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4323077","identity":"rs-4323077","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}