Clustering for Recommendation of Further Studies for Lectures at Politeknik Negeri Media Kreatif, Indonesia

Abstract

Professionally, lecturers must continue to develop themselves in their field of expertise and regularly update their knowledge and skills, in order to provide teaching that is up-to-date and relevant to the changing times. To improve the quality of lecturers, one of the strategies is to enhance their academic qualifications. In this regard, lecturers who meet the qualifications can pursue further studies at the doctoral level. The recommending authorities can consider various factors when evaluating the performance quality of a lecturer, including age, work experience, academic achievements and performance, specialization, lecturer-to-student ratio, availability and readiness of substitute lecturers, as well as the reputation and status of the university where the lecturer applies and is accepted. In this research, the clustering of lecturer performance data is done using the K-Means method to address the aforementioned issues. This allows for faster determination of recommendations as the process is automated. The analysis and evaluation of the clustering results are conducted by determining the quality of the clustering using the Silhouette Coefficient method. This research has successfully applied the K-Means Clustering method, providing 3 clusters of recommendations for further studies for Polimedia lecturers. The testing of this system using the Silhouette Coefficient obtained an average value of 0.78, indicating that the clustering results are in good condition.
 Keywords: K-Means, evaluating, resul