Mobile-Based Prediction Framework for Disease Detection Using Hybrid Data Mining Approach

Abstract

Early detection of life-threatening diseases is a major healthcare problem and early detection an immediate response is required to control diseases. Now a day’s mobile technology has become an important medium for doctors, patients and healthcare practitioners for seeking information related to health. In this study, we build a mobile-based disease detection system using hybrid data mining approach. The algorithms used are Sequential Minimal Optimization and Naïve Bayes and these classifiers were combined by using the voting ensemble technique. The proposed Framework enables us to enter the symptoms and then prediction model detects disease based on the symptom entered by the user. Proposed model is able to detect various diseases online. The proposed framework is evaluated on different disease datasets collected from UCI ML Repository. An experiment scheme was designed for checking the efficiency of the proposed hybrid model and the results depict that the proposed model is able to detect disease with average accuracy of 93.305 for different diseases.