<![CDATA[Abstract: The development of information technology in education is changing the way students construct and access knowledge, but differences in academic ability, motivation, discipline, and digital literacy often lead to learning disparities. This study grouped student characteristics using K Means Clustering and reviewed them from the perspective of computer science philosophy: ontology, epistemology, axiology, logic, and induction. The data from 120 students included academic scores, learning motivation, discipline, and digital literacy. After normalization, the number of clusters was determined using the Elbow and Silhouette methods, and the quality of the clusters was evaluated using the Davies–Bouldin Index. The findings indicate an optimal number of three clusters, with a Silhouette value of 0.466 and a DBI of 0.733, indicating fairly good and stable clustering. The three clusters describe: 1) highly motivated students with high digital literacy; 2) disciplined students with good academic performance but moderate digital skills; 3) low-motivated students with low digital literacy who require a personalized and empathetic learning approach. Ontologically, data is not just numbers, but the manifestation of students' digital existence in the modern learning space. Epistemologically, knowledge is formed inductively from students' interactions with technology and data. Axiologically, the clustering results support fairness in digital learning with an approach tailored to student characteristics. The dimensions of logic and induction show the clustering process as a scientific pattern of thinking from observation to meaningful rational generalization. The findings support a balance between algorithmic rationality and human values in digital education. Keyword: K-Means Clustering; Philosophy of Computer Science; Ontology, Epistemology; Axiology; Student Characteristics; Digital Learning. Abstrak: Perkembangan teknologi informasi di pendidikan mengubah cara siswa membangun dan mengakses pengetahuan, tetapi perbedaan kemampuan akademik, motivasi, kedisiplinan, dan literasi digital sering menimbulkan ketimpangan pembelajaran. Penelitian ini mengelompokkan karakteristik siswa dengan K Means Clustering dan meninjaunya melalui perspektif filsafat sains komputer: ontologi, epistemologi, aksiologi, logika, dan induksi. Data 120 siswa meliputi nilai akademik, motivasi belajar, kedisiplinan, dan literasi digital. Setelah normalisasi, jumlah klaster ditentukan lewat metode Elbow dan Silhouette, lalu kualitas klaster dievaluasi dengan Davies–Bouldin Index. Temuan menunjukkan jumlah klaster optimal tiga, dengan nilai Silhouette 0,466 dan DBI 0,733, mengindikasikan pengelompokan yang cukup baik dan stabil. Tiga klaster menggambarkan: 1) siswa bermotivasi dan berliterasi digital tinggi; 2) siswa disiplin dan berprestasi akademik baik, namun kemampuan digital sedang; 3) siswa bermotivasi dan literasi digital rendah yang memerlukan pendekatan pembelajaran personal dan empatik. Secara ontologis, data tidak sekadar angka, melainkan wujud eksistensi digital siswa dalam ruang belajar modern. Epistemologis, pengetahuan terbentuk secara induktif dari interaksi siswa dengan teknologi dan data. Aksiologis, hasil klasterisasi mendukung keadilan pembelajaran digital dengan pendekatan sesuai karakteristik siswa. Dimensi logika dan induksi menunjukkan proses klasterisasi sebagai pola berpikir ilmiah dari observasi menuju generalisasi rasional bermakna. Temuan mendukung keseimbangan antara rasionalitas algoritmik dan nilai kemanusiaan dalam pendidikan digital. Kata kunci: K-Means Clustering; Filsafat Sains Komputer; Ontologi, Epistemologi; Aksiologi; Karakteristik Siswa; Pembelajaran Digital.]]>
Copyrights © 2025
