Penerapan Data Mining Menggunakan Metode K-Means Untuk Analisis Pola Distribusi Zakat Pada Masjid Al-Ikhlas Karawaci

Abstract

The zakat distribution process at Al-Ikhlas Mosque Karawaci previously faced challenges regarding objectivity and effectiveness in allocating aid to eligible recipients. The absence of a measurable classification system led to a potential lack of precision in distributing community funds. Consequently, this research was conducted to develop a data analysis model capable of enhancing the accuracy and transparency of zakat distribution. The implemented solution involved the application of the K-Means Clustering algorithm within the Knowledge Discovery in Databases framework. This study contributed a data-driven zakat distribution model that integrated local criteria with Central Bureau of Statistics standards and Islamic law. The research methodology began with data processing of 296 mustahik from the 2024 Ramadan period. The variables utilized included zakat type, aid amount, number of dependents, economic conditions, and aggregate demographic data of the Karawaci District. The experimental phase involved comparing the performance of the K-Means algorithm against Hierarchical Clustering and Density-Based Spatial Clustering of Applications with Noise. The optimal number of clusters was determined using the Elbow method and Silhouette Score, which resulted in three main distribution groups. The results demonstrated distribution patterns of nineteen percent for the high category, 36 to 52% for the moderate category, and twenty-six to forty-five percent for the low category. Evaluations via Silhouette Score ranging from 0.35 to 0.58 and the Davies-Bouldin Index from 0.61 to 1.26 confirmed that the generated clusters were of high quality and consistent. In conclusion, the application of the clustering algorithm successfully established a strategic recommendation foundation for mosque administrators to improve zakat distribution precision