Comparative Analysis of Various Classification Models on Disease Symptom Prediction Dataset

Abstract

Data segregation is a vital task to label the class of data. Attributes are part of knowledge working in a math task. In this paper, we report the comparative study of various classifiers, i.e., K-Nearest Neighbor (K-NN), Decision Tree, Nave Bayes, Support Vector Machine (SVM) and Random Forest, analyze that which classifier works well beneath what conditions. For this purpose, medical datasets, i.e., UCI datasets have been selected. The performance of these classifiers have been evaluated in terms of Recall, Precision, Accuracy, and F1-Score. The accuracy for Decision Tree, K-NN, Nave Bayes, SVM and Random Forest, are observed to be 95.85%, 100%, 100%, 87.46% and 98.32%, respectively. The present study illustrates that the K-NN and Nave Bayes classifiers outperformed as compared to Decision Tree, SVM and Random Forest. Therefore, KNN and Nave Bayes classifiers can be used in automatic ailment and ascertaining diseases detection.