Analisis Pengelompokan Pola Pelanggaran Kode Etik Profesi TI Berdasarkan Karakteristik Insiden Siber Menggunakan Algoritma K-Means Clustering

Abstract

The rising intensity of cyber incidents in Indonesia is not merely a technical issue but also reflects failures in applying the professional code of ethics in information technology (IT), such as the obligation to avoid harm, preserve data confidentiality, and exercise professional competence responsibly. This study aims to group cyber incidents by their characteristics and then interpret the resulting clusters as indications of IT professional code-of-ethics violations. The K-Means Clustering algorithm was applied to the Cybersecurity Incident Dataset (Habeeb, 2024) using the CRISP-DM framework. The numerical variables analysed include financial loss, number of affected users, resolution time, severity score, and attack sophistication. The optimal number of clusters was determined by combining the Elbow method and the Silhouette coefficient. The analysis produced three distinct clusters, namely high-impact and sophisticated incidents, medium operational incidents, and high-volume low-impact incidents, with a Silhouette value of 0.53 indicating an adequate clustering structure. Mapping each cluster onto ethical principles shows that high-volume incidents are most associated with weak awareness and basic controls, whereas high-impact incidents are most associated with negligence of professional responsibility on critical systems. These findings can serve as a basis for more targeted mitigation prioritisation and professional ethics enforcement.