Crime Analysis & Prediction Using Machine Learning

Crime Analysis & Prediction Using Machine Learning

Prof.Nikita Joshi1,Sandip Chavan2 Yashraj Dalvi2, Ashutosh Dange2, Manthan Dimble2

1Student, Department of Computer Engineering, Savitribai Phule Pune University, Pune, India.

  Zeal College of Engineering, Pune, Maharashtra, India.

Abstract: - Understanding crime patterns is a crucial step in preventing and combating crime, as it enables the identification of prevalent crime types and their occurrence times in specific areas. Support Vector Machines (SVM), a machine learning algorithm, can be employed to classify different crime types and enhance prediction accuracy. The utilization of crime data from sources such as Kaggle can provide valuable insights into crime patterns, aiding law enforcement strategies and resource allocation. Analyzing crime data enables the identification of crime hotspots, facilitating targeted resource allocation. However, it is essential to acknowledge that crime data might not always reflect the true extent of crime in an area due to underreporting or lack of recording. Moreover, relying solely on data analysis may overlook the social and economic factors that contribute to crime. Therefore, while SVM and machine learning algorithms can assist in predicting and identifying crime patterns, adopting a multifaceted approach that incorporates community engagement, social services, and crime prevention initiatives is crucial to effectively address crime.

Key Words— Crime, Analyze, Crime patterns, Kaggle, Estimate, Support Vector Machine, Classification, Regression and Accuracy