Impact of Artificial Intelligence in Feature Selection Methods of Naïve Bayes Prediction Model

Abstract

: Naïve Bayes is a supervised and fast learning classification algorithm. It has a dominant role to play in the field of medical data mining. The use of data mining techniques in medical related research is known as Medical Data Mining. Data mining is a method of finding interesting trends and information from vast volumes of data. Healthcare sectors use data mining methods to increase their efficiency and quality of service. Naïve Bayes is one of the most prevalent algorithms working on the Bayes theorem. It fits well for a large number of datasets and provides better efficiency in classification tasks. Feature selection is a method of minimizing the number of input variables when creating a prediction model. It focuses primarily on eliminating non-informative or redundant predictors from the model. These methods become an indispensable part of the Naïve Bayes algorithm and enable our models to achieve higher accuracy in the prediction phase. This proposed work elucidates the importance of Artificial Intelligence in feature selection methods and also explains how these feature selection methods optimise the efficacy of Naïve Bayes' disease prediction algorithm.

Keywords: Medical Data Mining, Disease Prediction, Naïve Bayes, Artificial Intelligence