Performance Analysis of Naïve Bayes Correlation Models in Machine Learning

Abstract

Machine Learning(ML) usage ranges from individual research to developing predictive models for the prestigious organizations like reliance, Facebook, twitter and LinkedIn etc.,. The common point is usage of bulk data and applying some sort of the algorithms on that data and come up with predictive analytics or observing the useful patterns so as to reach the target customers and serve the public in a better way. The companies trying ML strategy to improve their business in drastic way and in many cases it has been proved.The current work focus on ML benefits and discussion of various algorithmic contexts like Naïve Bayes, Random Forest and Correlation analysis on certain data sets and our aim is to provide a basis of these algorithms and the usage models of algorithms along with some case studies. We believe that the work helps to understand the algorithms in simple way and helps the researchers to have some idea about the usage of the algorithms.To implement the algorithms R packages and methods we have used, R provides the importing the data and usage of the libraries related to algorithms and provides the plots so as to get the better understanding of the results.The significance of the work is describing the said algorithms along with research issues related to those aspects and publishing the results with analysis of the data sets. The outcome of the work is research issues related to the mentioned algorithms, result analysis and future scope of these works can be found.The algorithms naïve Bayes belongs to the category of supervised learning and comes under the category of classification techniques. Here supervised refers to the identified labels and expected outcome which can be achieved in the optimized way. The correlation analysis gives the idea about the kind of the relation between the entities which helps to keep track of the positive or negative correlation between the entities.