GIS-Integrated Machine Learning Framework for Predicting Traffic Accident Hotspots: A Spatial Analysis Approach

Jose C. Agoylo Jr.

© 2025 Jose C. Agoylo Jr., published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International. (CC BY-NC 4.0).

Citation Information: SAR Journal. Volume 8, Issue 2, Pages 176-180, ISSN 2619-9955, https://doi.org/10.18421/SAR82-08, June 2025.

Received: 06 June 2025.
Revised:  17 June 2025.
Accepted: 24 June 2025.
Published: 27 June 2025.

Abstract:

In city areas such as Metro Manila, Philippines, traffic accidents are the largest threat to urban growth and public safety. The K-Nearest Neighbors (KNN) method is used in this study's machine learning and GIS-based technique to predict the sites of traffic incidents. Based on spatial data and historical accident records, it integrates geographic and contextual factors to generate risk maps that identify locations likely to be accident-prone. There was room for improvement, as the predictive model had a 97% accuracy rate with good precision, but a somewhat lower recall. Urban planning initiatives, resource allocation, and traffic control methods can all benefit from the practical insights that GIS-based visualizations offer. Thus, the method suggests that combining machine learning and spatial analytics could improve road safety.

Keywords – GIS, machine learning, traffic accident prediction, K-Nearest Neighbors (KNN), spatial analysis

                   

                                                                      Full text PDF