K-Means Algorithm For Clustering Poverty Data in Bangka Belitung Island Province

Abstract

The Central Bureau of Statistics is a non-ministerial government institution that reports directly to the President. Based on data from The Central Bureau of Statistics in September 2019, the wealth rate in Indonesia was 9.22% and the number of indigent people in Indonesia reached 24.79 million. The poverty rate in the Bangka Belitung Islands Province was low compared to the national level. This is evidenced by 4.62% of people in Bangka Belitung Island Province were indigent people, which is lower than the national average of 9.22%. The data mining techniques using the K-Means Clustering method are used for this study. The research data was taken from the website of the BPS from 2014-2019 which consisted of 7 districts and/or cities with 3 variables. The variables used are the number of indigent people (in thousands), the average length of school education (years), and adjusted per capita expenditure (thousand rupiahs/year). All data is processed by Rapidminer and 3 clusters are carried out, namely: medium cluster level 0, high cluster level 1, and low cluster level 2. Cluster 0 contains districts/cities whose people have the longest average school time, high per capita expenditure, and a large number of indigent people. Cluster 1 contains districts/cities whose people have a short average school time, low per capita expenditure, a moderate number of indigent people. Cluster 2 contains districts/cities whose people have an average school time, moderate per capita expenditure, a small number of indigent people. Based on the result, the government can prioritize Kabupaten Bangka, Kabupaten Bangka Barat, Kabupaten Bangka Selatan in assisting, especially in the cost of education scholarships and social funds as well as other infrastructure improvements for the welfare of the inhabitants of Bangka Belitung Islands Province.

Influence of Motivation, Creativity, and Innovation on Food and Beverage Processing Industry Business Perfomance in Pangkalpinang

“…Based on research conducted by (Daud, Sudrajat, Maryani, & Effendi, 2021) obtained an inventory of MSME problems in the Bangka Belitung Islands Province, Pangkalpinang City is the capital of the Bangka Belitung Islands Province which has an area of 104,405 km2. When compared to the province, the area of this city is only 0.72 percent and is the smallest city/district area in the Province of the Bangka Belitung Islands (Sugianto & Bokings, 2021).…”

Evaluation of ATM Location Placement Using the K-Means Clustering in BNI Denpasar Regional Office

“…In this study, criteria and classifications or clusters can be determined using the K-Means method to determine the evaluation results of the placement of ATM locations. Here is a step-by-step process showing the K-Means clustering algorithm shown in Figure 2 (Sugianto & Bokings, 2021). of the above criteria determine the evaluation results by grouping the requirements using the clustering method and applying the K-Means algorithm in the study.…”

“…Based on the problems described previously, the purpose of this study is the strategic ATM placement evaluation system by implementing the K-Means method. Several previous studies related to the implementation of K-Means Clustering have also been successfully carried out, including for grouping sales data, decision-making of BLT recipients, beef production clustering, and clustering poverty data field (Halawa & Hamdani, 2019;Kusanti & Sutanto, 2021;Pandiangan, 2019;Sugianto & Bokings, 2021). Furthermore, this study also aims to test K-Means' effectiveness in increasing business potential and business benefits in evaluating BNI ATM placements.…”

Pengelompokan Kemiskinan di Indonesia Menggunakan Time Series Based Clustering

“…Penelitian terkait pengelompokkan kemiskinan kab/kota di pulau Jawa pernah dilakukan dan didapatkan hasil jumlah kelompok optimal sebanyak 2 dengan kriteria kelompok kemiskinan tinggi dan rendah [3]. Penelitian lain mengenai pengelompokan kasus kemiskinan pernah dilakukan dengan menggunakan metode k-means clustering menggunakan jarak Euclidean [4], [5]. Pengelompokan objek data yang mempertimbangkan series data dikenal juga sebagai time series based clustering.…”