Efficient Prediction of Crime Hotspots Using Deep Learning Approaches With Improved Accuracy

The main aim of this research is to predict crime hotspots from the given dataset using the algorithms Long short-term memory (LSTM) and recurrent neural networks (RNNs). Here LSTM is proposed and it will be compared over existing RNN. The result obtained will be improved accuracy rate. Introduction: Crime hotspot prediction is a subset of predictive policing, a field that uses data analysis and machine learning to forecast crime events. The core idea is to identify patterns and trends in historical crime data, allowing law enforcement to concentrate their resources on regions with a higher risk of criminal activity. This proactive approach aims to prevent crimes, enhance public safety, and optimize resource allocation. Materials andMethods: In this research two different algorithms namely Long short-term memory (LSTM) and recurrent neural networks (RNNs). will be used and prediction regarding crime hotspot is done based on accuracy. The tool used is python. Results and Discussion: The outcome of the recommended crime spot prediction is done and results are obtained interms of accuracy. The proposed LSM produces accuracy of about 92.17% which is very high compared to existing RNN which gives accuracy of 82.19%. Conclusion: This recommended research process result says proposed LSTM works better than existing RNN in predicting crime hotspots.