Crime Data Analysis and Prediction of Perpetrator Identity
A.Harshitha, P.Ramya, V.Yeshwanthi, V.Pranitha
Computer Science of Engineering, GRIET ,India
Department of CSE, GRIET, India
Dr K Butchi Raju, Professor, Department of CSE, GRIET, Hyderabad, Telangana
ABSTRACT
Crime is one of our society's most serious and distressing issues, and preventing it is a critical duty. Crime analysis is a method of finding and studying patterns and trends in crime that is done in a methodical manner. The purpose of this approach is to improve the efficiency of criminal justice systems. This model recognises criminal patterns based on inferences drawn from the crime scene and predicts the description of the offender who is most likely to have committed the crime. This project includes two primary components: crime analysis and perpetrator identity prediction.The Crime Analysis phase determines the number of unsolved crimes and investigates the impact of numerous characteristics such as year, month, and weapon on those crimes. The attackers' age, sex, and relationship with the victim are estimated during the prediction phase. The evidence gathered at the crime scene is used to make these predictions. Using methods like KNN Classifier and Neural Networks, the system guesses the perpetrator's description.It has undergone extensive training and testing using Kaggle Us open dataset and implemented using python.The implementation of this project results the speed up in the procedure time of the case solving based on the previous analyzed data which helps in the solving issues.So for the implementation is rather essential and that is the prime motive of the idea establishment. We use pycharm which supports constraints like matplotlib for future data visualization or the machine learning algorithms for analysis and prediction.To get the required result we consider sklearn, flask,html,sqlyog for requirements like webpage depiction and database for the execution.
Keywords
Multilinear Regression; K-Neighbors Classifier; Artificial Neural Networks; Kaggle Homicide Dataset;Python