Crime rate visualization with predictive analytics
Loading...
Date
2024-05
Journal Title
Journal ISSN
Volume Title
Publisher
Don Mariano Marcos Memorial State University - Mid La Union Campus
Abstract
This research utilized advanced machine learning models and artificial intelligence neural networks to develop and integrate a system for predicting and visualizing crime rate data in specific areas. Partnering with a government organization, the researchers carefully selected respondents to ensure comprehensive data collection. The development process presented numerous challenges, providing valuable learning experiences. Geographic Information System (GIS) graphs and heatmaps enhanced the system's efficacy and reliability. The study achieved the following objectives: identifying crime prediction indicators, training machine learning models, developing the crime prediction system using CRISP-DM methodology, and testing the system with the User Acceptance Test (UAT) and System Usability Scale (SUS) with six police personnel and four IT experts from DMMMSU MLUC. Significant findings included the identification of dataset indicators from San Juan Police Station data, the high accuracy of regression models, successful integration using the Flask API framework, and a 100% acceptability rate from UAT and 95% from SUS
Description
Full text
Keywords
Crime, Criminal investigation, Crime prevention, Reinforcement learning (Machine learning)[, Machine learning, Machine learning--Mathematical models, Machine learning--Statistical methods, Machine learning--Industrial applications, Machine learning--Technique, Machine learning--Graphic methods, Machine learning--Study and teaching, Simulated annealing (Mathematics), Computer algorithms
Citation
Buncab, L. M. P., Sandoval, J. S. F., Ocampo, J. L. T., & Almodovar, A. D. A. (2024). Crime rate visualization with predictive analytics. [Unpublished Undergraduate Thesis]. Don Mariano Marcos Memorial State University - Mid La Union Campus, City of San Fernando, La Union. Lakasa ti Sirib, DMMMSU Institutional Repository.