SENTIMENT ANALYSIS TO CLASSIFY THE SENTIMENT OF STUDENTS’ FEEDBACK USING THE CLASSIFICATION METHOD IN R AND POWER BI

Students’ feedback plays a huge important role in academic institutions to improve
teaching practices and learning processes. Quantitative feedback is given a lot of
attention in most universities as it is easier to be analysed. However, the qualitative
comments given by the students are often put aside. The reason is that due to its
unstructured messages form, it is difficult to analyse the feedback manually. To
counter this problem, text mining technique which utilize machine learning approach
will be implemented on students’ feedback to display the teaching assessment results.
Machine learning approach such as Naïve Bayes classifier, has been commonly used
for sentiment analysis to precisely and accurately describe the sentiment of an
individual review. By using the Naïve Bayes classifier, the outcomes of the teaching
assessment can be measured through defining the level of positive, neutral and
negative opinions. The implementing process of text mining technique from students’
feedback will be explained accordingly by using the statistical programming tool, R.
The teaching assessment results will be visualised and displayed in a dashboard
through the utilization of Power BI. This study is conducted with the aim to better
understand the sentiment analysis using Naïve Bayes classifier.
The main objective of this study is to use Naïve Bayes classifier to apply sentiment
analysis on the qualitative feedbacks obtained. The outcome of this study is classifying
the sentiments of the students’ qualitative feedbacks into positive, neutral and negative
and visualizing the sentiment results in a Power BI Dashboard.