Fuzzy-Based Predictive Analytics for Early Detection of Disease—A Machine Learning Approach

Abstract

Influenza is significant promising and re-emerging contagious disease, causing high morbidity and mortality throughout the world. Influenza also called swine flu, hoard flu, pig flu. The primary task of healthcare providers is providing diagnostic product services at low costs and to diagnose patients accurately. The use of data analytics and machine learning is a breakthrough technology that can have a significant impact on the healthcare field. Machine learning methods can be used for disease identification because they mainly apply to data themselves and give priority to outcomes of specific tasks. In this work, a framework of classification and regression trees (CART) algorithm is to create the fuzzy rules to employ for improved disease prediction. Validate the proposed method by choosing public medical datasets. Results on swine flu data demonstrate that our method distinguished the enhancement of the disease prediction accuracy. The implementation results exhibit that the aggregation of two methods such as fuzzy rule-based CART algorithm with irrelevant data elimination methods that could be useful in predicting disease. The proposed system can be helpful for healthcare providers as a healthcare analytical technique.