Web-Based Student Opinion Mining System Using Sentiment Analysis

Abstract

: Collecting feedback from a few students after the exams has been the norm in educational institutions. Forms are given to students to assess the course the lecturer has taught. The main purpose of developing student opinion mining system is to create a faster and easier method of collecting feedback from student, and also give lecturers and school administrators an easier way of analysing the feedback collected from students. The significance of this application is that it is less expensive and present a more confidential way of getting students opinion. The major tools used in developing this application are Python, Scikit learn, Textblob, Pandas and SQLite.. Django provides an in-built server that allows the application to run on the localhost.. In this project dataset gotten from online feedback form distributed to students was used for the sentiment analysi ,Chi-square was used for feature selection and the support vector machine algorithm was used for sentiment classification. The application will help the university administrators and lecturers to identify the strengths and weaknesses of the lecturer based on the textual evaluation made by the students. This paper studies different methods that could be used for learning sentiment from students’feedbacks. It utilized four techniques namely Naive Bayes, Complement Naive Bayes (CNB), Maximum Entropy and Support Vector Machine (SVM) on real-time students' feedback to identify sentiments.