Myanmar Text Classifier using Genetic Algorithm

Abstract

Text Classification is the task of automatically assigning a set of documents into certain categories (class or topics) from a predefined set. This also plays an important role in natural language processing and also crossroads between information retrieval and machine learning. The dramatic growth of text document in digital form news website makes the task of text classification more popular over last ten year. The application of this method can be found in spam filtering, question and answering, language identification. This book presents the idea of text classification process in term of using machine learning technique and illustrates how Myanmar news documents are classified by applying genetic algorithm. The applied system use Myanmar online news articles from Myanmar news website for the purpose of training and testing the system. Term Frequency Inverse Document Frequency (TFIDF) algorithm was used to select related feature according to their labelled documents which are also applied in many text mining methods.