An Overview of Methods for Feature Selection Based on Mutual Information for Stream Data Classification

Feature selection, the process of removing irrelevant features can be extremely useful in reducing the dimensionality of the data to be processed, in reducing execution time and improving predictive accuracy of the classifier. It is very hard to process over the large data set with respect to the time, memory and the computation cost. Because the bigger size data set are having the large number of the attributes to process, it is not necessary in many cases that all the features are relevant to get the knowledge about the domain. We want to decrease the number of features to process for classification.