Implementation of K-Nearest Neighbor Algorithm for Scientific Determination of Aid Recipients at STM Agape

Abstract

Providing assistance to underprivileged families is an important social effort to enhance community welfare; however, the selection of aid recipients often encounters problems such as subjectivity, unstructured data, and time inefficiency when conducted manually. This study aims to develop and evaluate a decision support system for determining aid recipients at STM Agape using the K-Nearest Neighbor (KNN) algorithm to improve accuracy and objectivity in the selection process. The research methodology employed a quantitative classification approach, where data were collected from families based on predefined criteria, including family income, number of dependents, housing conditions, and the occupation of the head of the household. The dataset was divided into training and testing data, and all attributes were normalized prior to processing. The KNN algorithm was applied using Euclidean distance to measure similarity between data instances, classifying each family into “eligible” or “ineligible” categories. The results indicate that the proposed system achieved higher classification accuracy and more consistent decision outcomes compared to manual selection methods. Additionally, the implementation of KNN reduced processing time and minimized subjective bias in determining aid recipients. These findings demonstrate that the KNN-based system is effective as a decision support tool, enabling STM Agape to distribute social assistance in a more targeted, objective, transparent, and efficient manner.