Analisis Clustering Data Penyandang Disabilitas Menggunakan Metode Agglomerative Hierarchical Clustering dan K-means

Abstract

Disability issues are still a major concern in society due to the discrimination often faced by people with disabilities. Many of them have abilities that are equal to individuals without physical limitations. Through this case study, this research aims to Cluster disability data by considering three types of disabilities: physical, visual and hearing, and hearing and speech using agglomerative hierachical Clustering and kmeans methods. This research was conducted by analyzing data from people with disabilities in 7 provinces in Indonesia. K-means to group data and agglomerative hierarchical Clustering as a centroid determinant in k-means. to enrich the results of data analysis, the EDA (Exploratory Data Analysis) process is used to identify outliers and anomalies. The results of the data analysis show that there are three main Clusters. The first Cluster has a high level of disability and includes 62 cities and districts, the second Cluster has a medium level of disability with 37 cities and districts, and the third Cluster has a low level of disability with 27 cities and districts. The best evaluation using the Davies Bouldin Index method resulted in two Clusters, indicating a better quality of Cluster division. The results of this study provide a better understanding of the distribution of disability in Indonesia, which can be used as a foundation to improve inclusion and accessibility for people with disabilities. Further recommendations can be made based on these findings to improve their situation in terms of employment and education.