Analyzing Student Perspectives on Learning Experience Using Latent Semantic Indexing Algorithm

The lecturer's evaluation by a student (EDOM) application is valuable for collecting anonymous feedback from students about their respective lecturers at the end of each semester. This feedback plays a crucial role in assessing teaching quality, identifying areas for enhancement, and promoting a constructive learning environment. This study addressed the challenge of effectively extracting and analyzing meaningful topics from a substantial dataset of 31, 600 student feedback records during the odd semester of the 2022/2023 academic year. Therefore, we proposed the adoption of the Latent Semantic Indexing (LSI) algorithm. This study also provides an in-depth description of the evaluation technique employed for assessing the coherence of the modeling results using the coherence score. Through experimental testing, we obtained notable coherence scores for each topic model. The optimal combination of parameters, a = 5 and ß= 20, resulted in the highest coherence score of 0.429, indicating the most proficient model for capturing latent structures within the data. The evaluation technique enabled quantifying the topic model's interpretability and consistency, significantly bolstering the credibility of the findings. The insights derived from the LSI analysis empower educational institutions to discern areas necessitating improvement, tailor pedagogical approaches, and cultivate a more effective and enriching learning environment for students.