Aspect-Based Sentiment Analysis of Students’ Feedback to Improve Teaching–Learning Process

Abstract

Nowadays, educational institutes and universities are showing interest to improve quality of education system by monitoring teacher’s teaching, student’s learning, and course analysis using feedbacks. Teaching–learning process of outcome-based education requires the maximum involvement of students, teachers, and other stakeholders to identify and evaluate different aspects of education. In today’s digitized world, a huge amount of opinions are expressed daily on teaching-related topics using different social media platforms. Posted statements from students and teachers can provide a potential source for evaluating the teaching–learning process. The management of this huge content is again a cumbersome and time-consuming job if it is done manually. It is also very difficult to extract opinions about different aspects of the written unstructured text. A huge amount of rationale and context are expressed daily on the social media platform. Sentiment analysis is a majorly used technique in finding the sentiment from the unstructured text. Sentimental analysis of online social media is related to minimizing the traditional way of collecting suggestion and feedback. Most of the work has been done to process user comments which are only to classify the positive or negative sentiment using lexicon-based or machine learning methods at document level. It is found that sentimental analysis is a largely underused tool in the educational context to find opinions on different aspects. In this paper, we have done aspect-based sentiment analysis using machine learning- and lexicon-based approaches.