Text Classification Based on Naive Bayes Algorithm with Feature Selection

Text Classification is the task to classify documents into predefined classes. It has become one of the key techniques for organizing information. Machine learning, a branch of artificial intelligence, has been used in text classification with better performance than rule based ones. But they mostly need lots of training samples in the processing, which not only brings heavy work for previous data collection, but also require a higher storage and computing resources during the processing. Naive Bayes is one of the most efficient and effective inductive learning algorithms and can get more accurate result in the large training sample set. To improve the performance, feature selection mechanisms are incorporated into naive bayes algorithm. Firstly, feature extraction techniques are applied to remove irrelevant and redundant features. After that, naive bayes classification algorithm is used to text classification. The experimental results have shown that this method keeps high classification accuracy.