A Comparative Study of Spam Detection in Social Networks Using Bayesian Classifier and Correlation Based Feature Subset Selection

Abstract

The article gives an overview of some of the most popular machine learning methods (Naïve Bayesian classifier, naïve Bayesian k-cross validation, naïve Bayesian info gain, Bayesian classification and Bayesian net with correlation based feature subset selection) and of their applicability to the problem of spam-filtering. Brief descriptions of the algorithms are presented, which are meant to be understandable by a reader not familiar with them before. Classification and clustering techniques in data mining are useful for a wide variety of real time applications dealing with large amount of data. Some of the application areas of data mining are text classification, medical diagnosis, intrusion detection systems etc. The Naive Bayesian Classifier technique is based on the Bayesian theorem and is particularly suited when the dimensionality of the inputs is high. Despite its simplicity, Naive Bayesian can often outperform more sophisticated classification methods. The approach is called “naïve” because it assumes the independence between the various attribute values. Naïve Bayesian classification can be viewed as both a descriptive and a predictive type of algorithm. The probabilities are descriptive are used to predict the class membership for a untrained data.