Özet:

Abstract Vector-borne diseases (VBDs) are one of the most serious human health issues, impacting millions of people each year in every corner of the globe. Multiple decision-making techniques are employed in this study to give a better diagnosis of VBDs. It assesses alternative illnesses with opposing symptoms. It is difficult to precisely define the weight of criteria and the ranking of alternatives (diseases) for each criterion. The proposed method is used to diagnose VBDs such as malaria, chikungunya, and dengue fever. In this paper, we proposed a prediction of VBD using various supervised machine learning classification algorithms. The Weka 3.7 machine learning framework has been used for the classification of data. The algorithms used, such as SVM, Naive Bayes, Adaboost, decision tree, ANN, etc., In extensive experimental analysis, we observed the SVM prediction had better detection and classification accuracy over the other machine-earning classes. For evaluation, we used 3000 records of patient data. The modified SVM (mSVM) achieves 100% accuracy for different cross validations.

Anahtar Kelimeler: