Text Classification Using Naïve Bayesian Classifier with Bigram

Abstract

Classification is a form of data analysis that can be used to extract models describing important data classes or to predict future data trends. Data classification is a two step process. This system is to study the Naïve Bayesian Classifier and to classify the class labels of data sets. In this system, classifier is built on the training data sets and tests the unknown datasets. And then, calculate the accuracy of classifier by using F1-Measure (F1-score). The Naïve Bayesian (NB) classifiers have been one of the most popular techniques as basis of many classification applications both theoretically and practically. Before the classifier is built, standard text documents are read, remove stop words and punctuations, stemming the words by using Porter Stemming Algorithm and then features are extracted by using Bigram probability based on keywords such as preprocessing step. The experiment is performed on IEEE and ACM standard documents, research documents. This system is determined the kind of document, such as medicine, computer, engineering and agriculture by using Naïve Bayesian Classifier.