Text Categorization (TC), also known as Text Classification, is the task of automatically classifying a set of text documents into different categories from a predefined set. TC uses several tools from Information Retrieval (IR) and Machine Learning (ML) and has received much attention in the last years from both researchers in the academia and industry developers. In this paper, we first categorize the documents using of three algorithm KNN, MLP, SVM based machine learning approach and two Data set: "Routers" and "Hamshahri" that the idea of combining multiple classifiers has been reviewed well and by combining Support vector machine (SVM), k-Nearest Neighbors (KNN) and Multi layer perceptron (MLP) algorithms,text and configuration of the "Hamshahri" for Persian documents and "Reuters" for English documents is classified. The used criteria for assessing and accuracy, along with experimental results on the Configure of the Hamshahri and Reuters by using of SVM, MLP and KNN algorithms indicated that the combination of algorithms and feature selection methods, while reducing the number of features, improves the efficiency and accuracy in the combining classifiers system.
