Fuzzy Multiple Rules based Machine Learning Diagnostic Prediction Model for Dengue disease in India
The aim of this research is developing a predictive model to diagnose dengue disease at early stages with an Indian perspective and to provides an efficient intelligent system to prevent an epidemic timely. For this purpose, author efforts are to collect results of dengue data reports from different Medical labs in Amritsar, Bhatinda and Hisar etc. This objective of this research is to design an efficient intelligent system for preventing dengue epidemic timely. Dengue is life-threating diseases and delays in treating this disease may cause of the high risk of loose life. Therefore, it becomes vital to detect dengue at early stages and diagnose this disease with higher accuracy. The author develops inferences rules using a multiple regression model for diagnosing dengue disease with a supply of input variables as a combination of symptoms and Medical Tests Like Ns1 IgM and IgG. After applying the regression model, the likelihood of occurring dengue disease is computed. After that, an inference engine is designed for diagnosing dengue at each level namely DENV 1, DENV2, DENV3 and DENV4. The propose Fuzzy Multiple Inference Rules-based Machine Learning Diagnostic Prediction Model for Dengue disease prediction model is developed with propose inference engine and apply machine learning algorithms Decision Tree, Random Forest and Support Vector machines which take inputs patient’s physical symptoms like fever, abdominal pain etc. and medical reports. These features are further converting these input variables in the dataset. The Fuzzy Multiple Rules-based Machine Learning Diagnostic Prediction Model for Dengue disease model contributes to diagnose dengue at an early stage, confirm dengue and serve dengue. The FMRPM Model had trained 70% and tested 30 % of 250 samples data of Dengue patients in the area of Amritsar from Jan 2019 to Dec 2019. From analytical and validation results, this model achieves 95 % accuracy for predicting dengue at each level.