Al-Powered Crime Hotspot Detection in Los Angeles Using CNN-LSTM and Geospatial Analysis

Chandrashekaraiah, Jaswanth (2025) Al-Powered Crime Hotspot Detection in Los Angeles Using CNN-LSTM and Geospatial Analysis. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (5MB) | Preview
Preview	PDF (Configuration Manual) Download (451kB) | Preview

Abstract

Urban crime is a threat to public safety, and it continues to be so even with high metropolitan cities such as Los Angeles. Forecasting crime risk accurately at a more spatial and temporal level can help law enforcement, policymakers, and the public engage in smart, proactive decision-making. This research proposes a deep-learning system for crime risk prediction, which integrates historical crime trends and contextual environmental factors in its projections on future crime risk levels. The geometric representation of the city is with regular H3 hexagonal grids, with each evaluated every month with respect to its spatial and temporal attributes. A hybrid model of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks is developed to extract features from historical sequences of crimes and from images within Google Street Views respectively. The interactive map, which provides intuitive understanding of crime distribution in the city, shows the predicted risk groups identified as low, medium, or high. The design is modular for possible future real-time realization. Ethical considerations about data sensitivity and fairness are integrated into the whole. This approach demonstrates the possibility of combining deep learning and geospatial encoding with urban imagery to advance understanding of crime risks in smart city applications.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Nagahamulla, Harshani UNSPECIFIED
Subjects:	H Social Sciences > HV Social pathology. Social and public welfare > Criminology > Crimes and Offences Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HT Communities. Classes. Races > Urban Sociology > City Planning H Social Sciences > HT Communities. Classes. Races > Urban Sociology
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	30 Jun 2026 17:39
Last Modified:	30 Jun 2026 17:39
URI:	https://norma.ncirl.ie/id/eprint/9415

Actions (login required)

View Item