164 PENGGUNAAN ANALISIS KLASTER K-MEANS DALAM PEMODELAN REGRESI SPASIAL PADA KASUS TUBERKULOSIS DI JAWA TIMUR TAHUN 2017 * Hardani Prisma Rizky 1 , Wara Pramesti 2 , and Gangga Anuraga 3 1 Program Studi Statistika, FMIPA, Universitas PGRI Adi Buana Surabaya, Indonesia, hardaniprisky@gmail.com 2 Program Studi Statistika, FMIPA, Universitas PGRI Adi Buana Surabaya, Indonesia, warapra@unipasby.ac.id 3 Program Studi Statistika, FMIPA, Universitas PGRI Adi Buana Surabaya, Indonesia, g.anuraga@unipasby.ac.id Indonesian Journal of Statistics and Its Applications (eISSN:2599-0802) Vol 4 No 1 (2020), 164 - 178 Copyright © 2020 Hardani Prisma Rizky, Wara Pramesti, and Gangga Anuraga. This is an open- access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Tuberculosis (TB) is a contagious infectious disease caused by the bacterium Mycobacterium tuberculosis which can attack various organs, especially the lungs. TB if left untreated or incomplete treatment can cause dangerous complications to death. East Java Province has the second-highest TB case after West Java Province. Therefore, we need statistical modeling to analyze the factors that influence TB in East Java Province. The data used in this study were sourced from data from BPS and East Java Provincial Health Offices in 38 districts/cities in East Java Province in 2017. Analysis of data using the OLS regression approach only looked at variable factors but was unable to know the effects of territory. So to overcome this, a spatial regression approach is used by comparing the weight of Queen Contiguity and the results of the k- means cluster analysis to obtain the best model. Based on the results of the analysis, the spatial aspects of the data have met the assumptions of spatial dependencies using the Moran's I test with a p-value of 0.000001295. The weighting matrix used is the k- means cluster weighting matrix k = 2. The test results obtained by the Spatial Autoregressive Moving Average (SARMA) model selected as the best model with the value of the deterrence coefficient (R2) and Akaike Info Criterion (AIC), 87.10% and 586.69. The factors that significantly influence the number of Tuberculosis patients in each district/city in East Java are population density (X2) and the number of healthy houses (X9). Keywords: moran's i, multiple linear regression, sarma, tuberculosis. * Received Aug 2019; Accepted Feb 2020; Published online on Feb 2020