Drug Prediction using the Machine Learning Techniques
Abstract
Now-a-days different types of micro-organisms are assaulting the human bodies due to the busy life style, and approaching the hospital for diagnosis is a challenging task. So, our research came-up with a solution which solved the problem by detecting the appropriate drug. This expert system has been developed by four machine learning algorithms which has secured 95% accuracy rate. The user needs to enter symptoms in python Graphical User interface (GUI), to get the disease and appropriate drug. The developed robust model saves the time by identify the disease within seconds. This is a novel technique for drug prediction in diseases. The max-wins voting approach technique is used in this expert system for predicting appropriate disease and drug. The dataset is of (4921 X 91) size and contains 41 unique symptoms. In this project these four algorithms such as K-Nearest Neighbors, Gaussian Naïve Bayes, Random Forest Classifier, and Decision Tree Classifier are used in identifying the disease. Our proposed system is beneficiary to drug stores, doctors and the common people by suggesting the appropriate drug which results in cost and time efficient.