Volume 19, No. 1, 2022

Lexicon-Based Approach For Sentiment Analysis To Student Feedback


Mrtdaa Mohammed Almosawi , Dr. Salma A. Mahmood

Abstract

Sentiment analysis is a tool used to determine the polarity (positive or negative) of texts using natural language processing techniques. It is one of the methods used to determine the student's satisfaction to the lecturer's performance and this process is common in educational institutions. In our research, we built a lexical-based sentiment analysis approach to determine the polarity of students' feedback. We collected a dataset by means of an open-ended electronic questionnaire. From this dataset we create a dictionary of opinions in the field of higher education that contains 2,217 words in both Iraqi dialect of southern Iraq and the modern Arabic language. The proposed sentimental analysis approach achieved an accuracy of 98%. We also applied machine learning algorithms Naive Bayes (NB), Support Vector Machine (SVM) and k-Nearest Neighbors (KNN) to achieve results (97%, 97%, 96%), respectively. The students' feedback was scored by 60% positive and 40% negative to the lecturer’s performance, for the collected dataset.


Pages: 6971-6989

Keywords: Iraqi Dialect, Modern Stander Arabic, Sentiment Analysis, Lexicon-based Sentiment, Machine Learning, NLP.

Full Text