Classification and analysis of users review using different classification techniques in intelligent e-learning system

Abstract

The internet comprised of a large number of data in the form of text, images, stickers, etc. which is also called as reviews or feedbacks created by users to share their expressions or knowledge. All those data may be in a different kind like positive, negative or neutral, and sometimes it may be in a single word or a single sentence or in document form. There are a few techniques, which are measured to provide better classifier like classification-support vector machine (SVM), Naïve Bayes (NB) and KNN. Using 'word tokenizer' in the techniques like (SVM, Naïve, KNN, J48, DT) it is compared with different results (accuracy, sensitivity, specificity, ppv and npv). It has been observed that after using various tokenizer in Weka tool (alphabetic tokenizer) has provided better results in measure, i.e., SVM (86.39%) comparing to techniques and specificity (83.77%) in average comparing to other measures.