Prediction of Swine Flu using a Hybrid Voting Algorithm

Abstract

Swine influenza disease is one of the causes that results in mortality in the modern society. The only way to eradicate this disease and save a person's life is to detect it at an early stage and cure it. Swine flu spreads easily by droplet transmission through cough, sneeze or even by touch. The main goal of this article is to find out which type of swine flu virus the patient suffering from by using Machine Learning algorithms. This article summarizes a prototype using data mining techniques, namely Random Forest, Support Vector Machine and a hybrid Voting Classifier Algorithm. And to determine which machine learning algorithm helps in finding accurate results in many fields. Using the patients’ symptoms as test data it was evaluated if the disease was prevailing or not. Hence, in this current article, it was explored to see if swine flu can be detected in its early stages so that it can be cured immediately after detection in an early stage. Along with the Swine Flu dataset of patients with symptoms classifiers such as Naive Bayes, k nearest neighbor, SVM, Random Forest and generated ROC curves were used to provide accurate predictions