<![CDATA[Technological developments, particularly in the field of machine learning, have had a significant impact on supporting data-driven decision making. One of the challenges faced in implementing PKH in Tiang Pumpung Kepungut Subdistrict, Musi Rawas Regency, is [A1] the process of determining aid recipients, which is still done manually. Data is still manually recorded into Excel based on data obtained during the population census. This often causes errors and mistakes during the aid distribution process. To overcome this problem, this study proposes the use of the Support Vector Machine algorithm in the PKH beneficiary classification process. Support Vector Machine is an effective classification method for handling complex and non-linear data with a high degree of accuracy. This study aims to develop a Support Vector Machine-based system to improve efficiency, accuracy, and transparency in the selection process for determining the eligibility of aid recipients. A total of 250 PKH data sets were successfully obtained. The data obtained or collected included several variables, namely name, age, number of family members, occupation, income, number of dependent children, and status. The data was then divided into two sets: 80% for training and 20% for testing from a total of 250 data points. After cleaning the data, the number of data points became 244, with 195 for training and 49 for testing. The results of this study showed that 39 families were eligible to receive assistance and 10 families were not eligible. It is hoped that the resulting system can serve as an innovative solution to support more targeted social assistance distribution in the region.]]>
Copyrights © 2025
