<![CDATA[Lecturer profiling based on scientific publications is a strategic component in managing human resources in higher education institutions. The manual process of identifying lecturer competencies often requires considerable time and may lead to inaccuracies. This study aims to develop an automated application for lecturer profiling and competency mapping to relevant courses using an unsupervised text similarity approach based on the Term Frequency–Inverse Document Frequency (TF-IDF) and Cosine Similarity methods. The application was developed using the Streamlit framework with integrated data from Google Scholar, SINTA, and Scopus. The evaluation involved 50 lecturers and 120 lecturer–course pairs, measured using accuracy, precision, recall, F1-score, response time, and usability metrics. The results show an accuracy of 85.3%, an F1-score of 0.853, an average response time of 2.3 seconds, and a usability score of 86.4, which falls into the excellent category. The system is capable of displaying interactive lecturer profiles, performing competency mapping to relevant courses, and generating automatic reports in PDF format. Therefore, this application effectively supports data-driven academic decision-making processes for assigning lecturers according to their areas of expertise.]]>
Copyrights © 2025
