Epidemiological Mapping Of Tuberculosis In South Sulawesi Using Local Indicators Of Spatial Association (LISA) And K-Means Clustering

Spatial statistics is a statistical approach that links data to the location of events. The most basic way to test whether data can be analyzed using spatial statistics is to find the spatial dependence. Local spatial dependence is tested using Local Indicators of Spatial Association (LISA). This research aims to use a form of LISA, Local Moran, to cluster and map epidemiological data, the number of tuberculosis (TB) cases in South Sulawesi. The novelty of this research is that the mapping of TB infectious disease in South Sulawesi was carried out using Local Moran, as well as clustering area using K-Means. The distribution pattern of TB cases in South Sulawesi tended to be clustered and the areas that had significant spatial dependency were Makassar, Maros and Takalar. The positive Moran value in Makassar shows that the characteristics of TB cases in Makassar tended to be similar to its neighbor. Meanwhile, the negative Moran values in Maros and Takalar indicates that the characteristics of TB cases in both areas were not similar to their neighbors. The result of K-Means shows that the areas with the highest number of TB cases in South Sulawesi were Bone, Gowa and Makassar.