Home / Current Issue / Paper 1707390
Crime Prediction and Analysis using CNN & RNN
Subject area: Science,Engineering and Technology · Area of research: Computer Science and Engineering
Abstract
Crime prediction and analysis play a vital role in enhancing public safety and optimizing law enforcement efforts. This study explores deep learning-based approaches, integrating Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks for effective crime forecasting and analysis. The proposed framework leverages the temporal strengths of RNNs and LSTMs alongside the spatial feature extraction capabilities of CNNs to analyze large-scale crime datasets. RNNs and LSTMs handle time-series data to predict future crime trends, while CNNs perform geospatial analysis to identify crime distribution patterns across regions. The hybrid model processes both structured data (e.g., dates, times, locations) and unstructured data (e.g., crime descriptions) to enhance predictive accuracy. Experimental results demonstrate its ability to detect crime hotspots, predict crime categories, and uncover hidden trends, offering actionable insights for law enforcement and policymakers. This study highlights the potential of deep learning in tackling complex, dynamic challenges such as crime prediction, contributing to smarter and safer cities. Future work could incorporate real-time data streams and assess the ethical considerations of deploying such models in decision-making systems
References
[1] Scarfone, K., CMell, P. (2007). Guide to IntrusionDetection and PreventionSystems (IDPS). National Institute of Standardsand Technology (NIST).
[2] Northcutt. S. C Novak J. (2002). Network Intrusion Detection: An Analyst Handbook. New Riders Publishing.
[3] Crime Forecasting: A ML and computer vision approach to crime prediction Neil Shah Nandish Bhagat & Manan Shah.
[4] Crime Prediction model using Deep neural Networks: Soon Ae Chun,Venkata Avinash Pataru.
[5] Data Mining and Region Prediction Based on Crime using Random Forest: Dewan Mamun Raza, Debasish Bhattacharjee Victor.
[6] -./<=?LMOZ[efghrstuvwyz|}~€�ñåÙ̶©Â¶©Â¶¡˜�ˆ¡˜}odYdYdYdNYNhX1±6�H*OJQJhJ:ö6�H*OJQJh9wþ6�H*OJQJh^~ h^~ 6�H*OJQJh^~ h^~ OJQJhJ:öOJQJhX1±OJQJhX1±H*OJQJh|ÓOJQJh^~ h^~ OJPJQJh^~ h^~ H*OJQJh|ÓOJPJQJhJ:öhJ:öOJPJQJh^~ CJ(OJQJaJ(hEq5CJ(OJQJaJ(hEq5hEq5CJ(OJQJaJ(./tÖ×ØÙ(ÝüýòòäÚų³¨�•�~~$„Cd]„Ca$gd‘…$a$gd¤=u &Fgd¤=u $da$gd¤=u $da$gd|Ó„ñÿ„ dð¤]„ñÿ^„ gd^~ $„ñÿ„ dð¤]„ñÿ^„ a$gd^~ $¤a$gdJ:ö [Some characters in this reference could not be displayed correctly — please refer to the published PDF for the full reference.]
[7] $„öÿ¤`„öÿa$gdJ:ö$dݤa$gdEq5 [Some characters in this reference could not be displayed correctly — please refer to the published PDF for the full reference.]
How to cite this paper
@article{1707390,
author = {Yashas B, Srinidheesh M, Chanackya J, Sacheet Kumar},
title = {Crime Prediction and Analysis using CNN & RNN},
journal = {Iconic Research And Engineering Journals},
year = {2025},
volume = {8},
number = {9},
pages = {175-179},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1707390.pdf},
abstract = {Crime prediction and analysis play a vital role in enhancing public safety and optimizing law enforcement efforts. This study explores deep learning-based approaches, integrating Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks for effective crime forecasting and analysis. The proposed framework leverages the temporal strengths of RNNs and LSTMs alongside the spatial feature extraction capabilities of CNNs to analyze large-scale crime datasets. RNNs and LSTMs handle time-series data to predict future crime trends, while CNNs perform geospatial analysis to identify crime distribution patterns across regions. The hybrid model processes both structured data (e.g., dates, times, locations) and unstructured data (e.g., crime descriptions) to enhance predictive accuracy. Experimental results demonstrate its ability to detect crime hotspots, predict crime categories, and uncover hidden trends, offering actionable insights for law enforcement and policymakers. This study highlights the potential of deep learning in tackling complex, dynamic challenges such as crime prediction, contributing to smarter and safer cities. Future work could incorporate real-time data streams and assess the ethical considerations of deploying such models in decision-making systems},
month = {March},
}