Arabic Text Categorization using k-nearest neighbour, Decision Trees (C4.5) and Rocchio Classifier: A Comparative Study

Abstract

No doubt that text classification is an important research area in information retrieval. In fact there are many researches about text classification in English language. A few researchers in general talk about text classification using Arabic data set. This research applies three well known classification algorithm. Algorithm applied are K-Nearest neighbour (K-NN), C4.5 and Rocchio algorithm. These well-known algorithms are applied on in-house collected Arabic data set. Data set used consists from 1400 documents belongs to 8 categories. Results show that precision and recall values using Rocchio classifier and K-NN are better than C4.5. This research makes a comparative study between mentioned algorithms. Also this study used a fixed number of documents for all categories of documents in training and testing phase.