An Effective Feature Selection method using Monte Carlo Search

Abstract

Feature selection is the challenging problem in the field of machine learning. The task is to identify the optimal feature subset by eliminating the redundant and irrelevant features from the dataset. The problem becomes more complicated when dealing with high-dimensional datasets. In this paper, we propose the novel technique based on Monte Carlo Tree Search (MCTS) to find the best feature subset to classify the dataset in hand. The effectiveness and validity of the proposed method is demonstrated by experimenting on many real world datasets.