K-Modes Analysis with Validation of the DBI in Grouping Provinces in Indonesia based on Indicators of Poor Households

Syifa Azahra
Zilrahmi
Dodi Vionanda
Fadhilah Fitri

UNP Journal of Statistics and Data Science
Vol. 2 No. 2 (2024): UNP Journal of Statistics and Data Science

Poverty is the most pressing social problem in Indonesia. Efforts to alleviate poverty are to group provinces in Indonesia based on indicators of poor households using the K-modes algorithm. The data used is data from the 2017 Indonesian Demographic and Health Survey (IDHS) on the Household List. The analysis includes data noise detection, data clustering using K-Modes algorithm, and cluster validation with Davies Bouildin Index (DBI). Based on the clustering that has been done, two clusters are obtained, where cluster 1 consists of 26 provinces and cluster 2 consists of 8 provinces. cluster 1 is a cluster that fulfills 9 indicators of poor households and cluster 2 only a few indicators of poor households. So that the government can prioritize these 8 provinces to overcome poverty in Indonesia. For the DBI value obtained is 1.89 which means that 2 clusters are already well used in the algorithm.