<![CDATA[Topic modeling is one of the text mining techniques that can be used to explore research themes in a collection of scientific documents. This study aims to identify and compare topic trends in SINTA-indexed national journal publications with student articles published in the ITG Algorithm Journal in the field of informatics and computers. The research data consisted of article abstracts that were analyzed through text preprocessing and text representation using bag-of-words, then modeled using Latent Dirichlet Allocation (LDA). The optimal number of topics was determined based on the coherence score, visualized using pyLDAvis, and labeled with the help of ChatGPT to clarify the interpretation. The results show that national journals emphasize application and information system development, while the ITG Algorithm Journal tends to address cutting-edge issues such as machine learning and data science. These findings contribute to mapping the development of information system research and can serve as a reference for formulating research policy directions at the local and national levels.]]>
Copyrights © 2025
