IMPLEMENTASI K-MEANS DAN K-MEDOIDS DALAM PENGELOMPOKAN WILAYAH POTENSIAL PRODUKSI DAGING AYAM

Abstract

Livestock is the main sector in the effort to fulfill food needs for people in Indonesia and has the potential to maintain the availability of animal food. Guidance and socialization to provide information and knowledge in the field of food production from animals, especially in areas with low levels of chicken meat production need to be done. The research objectives were the use of the K-Means and K-Medoids algorithms for grouping chicken meat production areas in the province of West Java and the use of the Davies Bouldin Index (DBI) value in choosing the best algorithm. The application of K-Means and K-Medoids was carried out through the data mining process phase, namely data collection, data preprocessing, data mining implementation, evaluation of the number of clusters, determination of the best algorithm, and clustering results. The K-Means algorithm with 5 clusters can optimally classify potential areas for chicken meat production in West Java province with a DBI value of 0.273. The results of clustering can be used in business processes related to information on the amount of chicken meat production in the West Java region as a reference in the pattern of guidance to increase animal food production, develop chicken farming potential, and develop animal feed distribution potential.

Keywords: clustering, chicken meat k-means, data mining, k-medoids