Forecasting the Rate of Crimes Committed Against Women Using Machine Learning Approach and Data Mining Techniques

Prediction of the probability of crime in any region is a serious problem, especially in developing countries. Several factors can influence the crime rate at any location. There have been no effective solutions or dedicated research studies on the increasing crime cases, especially among women. Thus, we have trained several machine learning algorithms and a neural network to come up with the best probability prediction model for different crime types. The data used was gathered from 50 districts in Madhya Pradesh, India. The trained models were evaluated based on their accuracy and f1-score in predicting the probability of different types of crimes.