<![CDATA[Credit risk assessment is a major challenge in the management of savings and loan cooperatives, especially when traditional methods are often affected by subjective biases and limitations in analyzing data systematically. This research aims to apply the K-Nearest Neighbors (KNN) algorithm in predicting credit risk accurately and efficiently, with a focus on analyzing borrowers' demographic features and credit history. The research methodology involved primary data collection from savings and loan cooperatives, descriptive statistical analysis, and performance testing of the KNN model using evaluation metrics such as accuracy, precision, recall, and F1-score. The analysis showed that the KNN algorithm achieved an accuracy of 85%, with high recall, indicating the model's ability to detect credit risk consistently. This research makes theoretical contributions by strengthening evidence of the effectiveness of machine learning in financial risk management as well as practical implications in the form of increased efficiency and objectivity in credit decision making. For broader generalization, future research is recommended to use more diverse datasets and explore other more complex algorithms. In addition, ethical aspects such as algorithm transparency and personal data protection should be the main concerns in field implementation.]]>
Copyrights © 2024
