A Comparison of K-Means and Fuzzy C-Means Clustering Algorithms for Clustering the Spread of Tuberculosis (TB) in the Lungs

Abstract

Tuberculosis (TB) is an airborne infectious disease that affects people of all ages, including infants, children, teenagers and the elderly. This disease is prevalent in different areas of Indragiri Hilir Regency, so it is important to identify and group the areas that are the focus of its spread. The purpose of this study is to help hospitals organize training in areas where tuberculosis is common. This study uses a data mining method with grouping techniques of K-Means and Fuzzy C-Means algorithms based on patient data from Puri Husada Tembilahan Hospital from 2020 to 2023. After several experiments, the results were evaluated with DBI, which showed that K- Means gave the best validity with a value of 0.9146. Which shows that the areas with high risk of TB are Tembilahans aged 55-64 who have been diagnosed with complicated TB. This method was then applied to the TB group information system of Puri Husada Tembilahan District Hospital in the hope that it could help the hospital reduce the spread of the disease in the affected area.Keywords: DBI, fuzzy c-means, clustering, k-means, tuberculosis.