Crimes have a negative effect on any society, both socially and economically. Law enforcement agencies face many challenges when trying to prevent crime. We offer a Criminal Data Analytics Platform (CDAP) to help law enforcement perform descriptive, predictive, and prescriptive analytics on criminal data. CDAP has a modular architecture where each component is built separately from each other. CDAP also supports plug-ins which allow for future functionality extensions. it can then analyze it, train models, and then visualize the data. CDAP also combines census data with crime data to get a more comprehensive analysis of crime and its impact on society. Additionally, with the combination of censusand crime data, CDAP provides process re- engineering steps to optimize the allocation of police resources. We demonstrate the utility of the platform by visualizing t and emotional spaces and relationships in a series of real-world crime datasets.The platform's predictive capabilities are demonstrated by predicting crime categories, for which a machine learning approach is used. Nave Bayesian, Random Forest Classifier and Multilayer Perceptron Network classification algorithms are provided to build a model. Optimized police district boundary identificationand patrol assignment are used to demonstrate the tool's prescriptive analytical capabilities. A heuristic-based clustering approach was adopted to define the boundaries of the police districts so that the identified districts have an equal population distribution with a compact shape. Theresulting districts are then scored for inequality and compactness of the population using the Gini coefficient and the isoperimetric quotient.

clustering, Machine Learning Algorithm,CDAP.