TOPICAL NEWS CLASSIFICATION USING MACHINE LEARNING TECHNIQUES

ABSTRACT

News is information that is presented through print, broadcast, Internet, or from mouth to mouth. For the ease of news, we classify news based on different category to help users to find relevant news rapidly. This classification results in the use of classifier engine to split any news into the respective category. This research employs the use of Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB) and Support Vector Machine (SVM) to classify topival news. The aim of this research is to develop a framework to categorize news topics in various categories and the objectives of this work are to pre-process the data using Term Frequency Inverse Document Frequency (tf-idf) and Bag of Words (BoW) which is suitable for the input to the classifier, apply Machine Learning (ML) techniques on pre-processed data, evaluate the performance of the machine learning classifiers on the pre-processed data and obtain the highest accuracy of the machine learning classifiers suitable on the pre-processed data. The finding shows SVM is a better classifier than NB, RF and DT using TFIDF while NB is a better classifier than SVM, RF and DT using BoW. Also, SVM is a better classifier using large datasets while NB thrives better with a smaller datasets.

Keywords: ML Classifiers, BoW, TF-IDF, NB,SVM.