Supervised Hybrid Framework for Addressing Word Sense Ambiguity with Special Focus on English-to- Sanskrit Machine Translation

One of the most important application of Natural Language Processing is Machine Translation (MT). It is an automated process of translation through a computer system. Machine Learning (ML) is one of the recent methods used in MT, and it has become a buzzword in the area of research over the last numerous years. Ambiguity is a major challenge in MT. ML has given promising results in terms of system learning and predicting results. The text classification technique in Machine Learning is considered as one of the most important methods to resolve Word Sense Disambiguation (WSD). The role of Dataset both as Training and Test data is important to predict the required results. We have taken data size of 2000 sentences which is further divided as training and test data. The dataset plays a pivotal role in validating the output of the system. We have also done an analysis on supervised machine learning text classification algorithms namely Naïve Bayes’, Decision Tree, Support Vector Machine (SVM), K-nearest Neighbor (KNN), Neural Network, Logistic Regression and Random Forest. The accuracy of the given algorithms range between sixty-eight to eighty-four percent. Further, we have also done analysis on proposed a “hybrid model” for prediction of POS ambiguity. In the proposed model, we have combined the Naïve Bayes’, Support Vector Machine and Decision Tree algorithms to achieve better results. The proposed model has reported an accuracy of eighty-five percent. The accuracy of the algorithms and model is tested using tenfold cross-validation method. The model has also reported high precision, recall and F-score in comparison to all supervised machine learning classification algorithms. The correctness of algorithm and model is tested in terms of the total number of correct POS predicted. The algorithms and proposed model are analyzed with the help of machine learning tool developed named as “AmbiF”.