Sentiment Analysis and Word Cloud of Teachers’ Evaluations Using R Programming Language
Catleen Glo Feliciano · Isabela State University Linker Journal of Engineering Computing and Technology · 2025
Faculty evaluation is essential for understanding students' perceptions and feedback to improve the employment of teaching strategies. With the use of vast-scale textual feedback, in an efficient manner, sentiment analysis was used as a tool for analyzing textual semantics in a structured way that could help facilitate understanding of what students think. Using the datasets of students' feedback from faculty evaluation from A.Y. 2019-2020 to A.Y. 2024-2025 for sentiment analysis using R programming, this study utilized Natural Language Processing (NLP). Data preprocessing, word cloud creation, and sentiment classification using code were employed to systematically extract prevalent themes, classify sentiments, and examine faculty performance. The approach comprises several processes, such as data preprocessing, word cloud generation, and sentiment classification, which are used to classify sentiments that follow an organized topic extraction and present useful insights about teacher performance. In fact, according to the data, students are overwhelmingly positive, with a deep appreciation for teachers who are helpful, efficient, and supportive in their teaching style and approach. The result also reflects how much students value the hard work that their teachers do, such as the top positive word is kind (mabait). Though they are less common, unfavorable opinions do draw attention to the areas in which students struggle, especially when it comes to their academic performance. While there are terminologies that reflect occasional problems in the classroom, where the top negative words are limit and hardship (hirap), it was noted that certain students struggle with their tasks. The results highlight how crucial it is to have a welcoming and interesting learning environment. Teachers may reinforce their strengths and highlight areas for growth by using sentiment analysis to get insightful information about student responses. Finally, by ensuring a well-rounded, efficient, and student-centered teaching approach for students pursuing a Bachelor of Science in Computer Science, this study offers a data-driven method of improving the learning experience.