Publications

Human life experiences a radical change due to the availability of all kinds of information on the World Wide Web. This phenomenon makes the life of each individual easier and better than before. This flooding of information also contains huge amount of people’s opinions about their used products or services. The massive flow of information in terms of opinions and views need to be analyzed to know the efficacy of the products and/or services. The subjective expressions, which are known as opinions describes the peoples’ feelings or sentiments towards the service or product. The opinion can be either positive or negative, depending on the evocation that the product or service has created. Similarly, in the realms of education, initiatives are being taken by Govt. Organizations as well as private organizations to know the effectiveness of the contemporary teaching-learning process. The present undertaking of the investigation is to evaluate the academic transaction by eliciting the students’ subjective expressions and analyzing the same as their feedback. The scope of this paper is to analyze the student’s feedback, which covers the teaching quality, style of delivery by the faculty, the components of the course, and the overall ambiance in which the academic activities are conducted. The present study investigates the applicability of different supervised machine learning approaches for sentiment analysis from students’ subjective feedback. The chosen supervised machine learning approaches are Naive Bayes (NB), N-gram, Support Vector Machine (SVM), and Maximum Entropy (ME). These four approaches are applied only for binary sentiment classifications and to obtain a comparative analysis that would help to build a superior teaching and learning process.