Poverty describes a condition of lack of ownership and low income, or in more detail describes a condition that basic human needs cannot be fulfilled, namely food, shelter, and clothing. In the last ten years, Central Java's poverty reduction performance has had its ups and downs, with rural poverty still dominating. The purpose of this research is to conduct a mapping analysis in the form of clusters on the number of poverty levels in districts or cities in the province of Central Java using artificial intelligence techniques. Given that Central Java is the third most populous province after West Java and East Java. This needs to be done in order to obtain a macro picture of the poverty level over the last few years through regional mapping. The dataset used is sourced from the Central Java Statistics Agency (BPS) website on the subject of the number of poor people (thousands of people) in 2017-2019. The solution given in conducting mapping in the form of clusters is the K-Medoids method which is part of clustering data mining. The number of clusters used are high and low clusters in mapping the number of poverty levels. The mapping analysis process uses the help of RapidMiner software. The results showed that 6 provinces (17%) were in the high cluster and 29 provinces (83%) were in the low cluster. The final centroid values for each cluster are {293.2, 309.2, 343.5} in the high cluster (cluster_1) and {18.6, 19.4, 20.1} in the low cluster (cluster_0). The results of the mapping can be useful information for tackling the poor where the high cluster (cluster_1) is a priority for the government in the province of Central Java, namely Cilacap Regency, Banyumas Regency, Kebumen Regency, Grobogan Regency, Pemalang Regency, Brebes Regency

Poverty describes the absence of property and poor income or the circumstance that the food, shelter, and clothing needs cannot be met. The performance of Central Java in poverty reduction has risen and declined during the last decade. This research aims to perform mapping analyses utilizing artificial intelligence techniques as clusters on the number of poverty levels in Central Java districts or cities. Since Central Java is after West Java and East Java is the third most populous province, this was necessary to achieve in the last few years through regional mapping of a macro-picture of the poverty level. The dataset used is from the statistics agency website on the number of poor people (millennia) in 2017–2019. The data used are from the Central Java Statistical Agency. The way to map the clusters is using the k-medoids method which is part of data clustering. The number of clusters utilized for mapping poverty levels is high and low. The results showed six provinces (17%) in the high and 29 (83%) in the low. In the high cluster (cluster 1) and in the low cluster (cluster 1) and {18.6, 19.4, 20.1} the final centroid values for each cluster were {293.2, 309.2, 343.5}. The results of mapping can help address the poor in places in which the high cluster (cluster 1), Cilacap District, Banyumas District, Kebumen District, Grobogan District, Pemalang District, and Brebes District are a priority of the government in Central Java province. Keywords: K-Medoids, clustering, poverty