<![CDATA[ABSTRAK Ketimpangan kualitas laboratorium jaringan pada SMK Teknik Komputer dan Jaringan (TKJ) di Sulawesi Utara menghambat upaya pemerataan kualitas pendidikan vokasi. Penelitian ini bertujuan menganalisis dan mengelompokkan tingkat kesiapan laboratorium jaringan pada 37 SMK di wilayah Manado, Tomohon, Minahasa, dan Minahasa Utara menggunakan metode K-Means Clustering. Data dikumpulkan melalui kuesioner terhadap 228 responden (kepala program, guru produktif, pengelola laboratorium, dan siswa) serta observasi lapangan. Tahapan penelitian mengikuti kerangka Knowledge Discovery in Database (KDD): seleksi data, data cleaning, transformasi, data mining, hingga interpretasi hasil. Penentuan jumlah cluster optimal dilakukan menggunakan Elbow Method, sedangkan kualitas clustering dievaluasi menggunakan Silhouette Coefficient. Hasil analisis mengidentifikasi tiga cluster kesiapan: Cluster 2 (Sangat Siap) sebanyak 22 sekolah (59,5%), Cluster 1 (Siap) sebanyak 12 sekolah (32,4%), dan Cluster 0 (Kurang Siap) sebanyak 3 sekolah (8,1%). Evaluasi dengan Silhouette Score menghasilkan nilai 0,1804 yang menunjukkan cluster terbentuk secara positif dan dapat membedakan karakteristik antar kelompok. Temuan ini memberikan dasar yang objektif bagi Dinas Pendidikan Provinsi Sulawesi Utara untuk menetapkan prioritas pengembangan fasilitas laboratorium secara lebih tepat sasaran. ABSTRACT The disparity in the quality of network laboratories in Computer and Network Engineering (TKJ) vocational high schools (SMK) in North Sulawesi hinders efforts to equalize vocational education quality. This study aims to analyze and classify the network laboratory readiness levels of 37 SMKs in Manado, Tomohon, Minahasa, and North Minahasa using K-Means Clustering. Data were collected via questionnaires from 228 respondents (program heads, productive teachers, laboratory managers, and students) alongside direct field observations. The research followed the Knowledge Discovery in Database (KDD) framework: data selection, data cleaning, transformation, data mining, and result interpretation. The optimal number of clusters was determined using the Elbow Method, while clustering quality was evaluated using the Silhouette Coefficient. The analysis identified three readiness clusters: Cluster 2 (Very Ready) with 22 schools (59.5%), Cluster 1 (Ready) with 12 schools (32.4%), and Cluster 0 (Not Ready) with 3 schools (8.1%). The Silhouette Score of 0.1804 indicates that the clusters were formed positively and can distinguish inter-group characteristics. These findings provide an objective basis for the North Sulawesi Provincial Education Office to prioritize laboratory facility development more effectively.]]>
