Klasterisasi Data Penerima Bantuan Langsung Tunai Menggunakan Algoritma K-Means

Abstract

Increasing population and unequal distribution of population even with conditions of varying poverty levels need to be the center of attention and proper handling. In Pelangsian Village, there were 202 residents who received BLTD in 2021. The existence of a quota of beneficiaries and the number of recipients' conditions that were not suitable often became an obstacle in determining beneficiaries. So that from the data obtained in this study it is necessary to do clustering. Clustering results can be used to find out if the population receiving BLTD meets predetermined criteria. so that it can further assist the government in seeing the categories of people who are really entitled to get this assistance. Data clustering can be done using algorithms in data mining. The algorithm used in the data clustering of Pelangsian villagers in this study is the K-Means algorithm. The research methodology was carried out in several stages, such as problem selection, data collection, data preprocessing, data mining algorithm selection, results evaluation, and results interpretation. Clustering is done by forming 2 data clusters. Before the data is clustered, 202 records need to be preprocessed so that it is found that there are 196 valid data records that can be processed according to research needs. The results of data processing are done by clustering the data into 2 groups. Clustering uses the K-Means algorithm by determining the value of K = 2 so that it is obtained that cluster0 has 115 residents and cluster1 has 81 residents. Algorithm performance testing shows that the K-Means Algorithm obtains a Devies-Bouldin value of -0.794. With a Davies-Bouldin-0.794 value, it can be said that the performance of the clustering algorithm is quite good.