Reliable Disease Prediction System Based on Association Rule Mining and Pyramid Data Structure

Abstract Data mining rules the world of data, as it is the base for analysing and relating the data with each other. Now-a-days, association rule mining concepts are applicable to almost all domains and the healthcare domain is in strong need of that. Taking this into account, this work attempts to propose a reliable disease prediction system with the help of association rule mining and pyramid data structure. This work processes a symptom dataset which is comprised of twelve health attributes. The health attributes with respect to four different diseases are clustered independently and are organised with the help of pyramid data structure. The process of clustering is carried out by Generalized Hierarchical Fuzzy C Means (GHFCM) and the strong association rules are built for predicting the disease. Finally, the effectiveness of disease prediction is evaluated with respect to the standard performance measures such as accuracy, precision, recall, F-measure and time consumption analysis. The performance of the proposed approach is observed to be satisfactory, which when compared to the existing techniques.