JJEM: Special Edition 2 (December 2024)
JJEM: Special Edition 2 (December 2024)
2024-12-08
Predictive Modeling of Criminal Activity Hotspots
Maithili Bhat, Arun Kumar K L
In our society the major cause in disturbing the peace of the country is Crime. The rate of crimes in society is an ever-increasing problem. Crime is a deliberate act of physical or physiological harm, damage or loss. Early prediction of the crime hotspot may help the law enforcement to take necessary action; in this regard the researchers are working to identify the crime zone which is affecting the peace and morality of the society using machine learning. Crime prediction helps the law enforcement committee to identify the most crime prone subjected areas and its pattern of crime. The previously recorded datasets helps in finding the pattern of the crime and speculate which are the high risk areas with the style of crime. This can be done using a KNN algorithm has potential to find spatial and temporal patter which makes it simple and transparency. KNN is ease to deploy with minimum training dataset. Our approach will help to find the crimes and its awareness in the country. Our work gives the 60-80% accurate results and can predict the crime hotspot. This work shows the potentiality of crime clustering, providing a solid foundation for next studies and useful applications in crime prevention techniques.
K-means, One-hot encoding, label Encoder, Elbow method, K-NN.