Crime Prediction and Analysis using Machine Learning
G Divya, Naman Saxena, Akshansh Sharma
Abstract
The escalation of criminal cases in India has been a pressing issue, resulting in a mounting backlog of pending cases. This continuous surge in criminal activities has posed a significant challenge in the classification and resolution of these cases. Recognizing patterns of criminal activity within specific regions is instrumental in preventing and mitigating crimes. Law enforcement agencies can enhance their performance by gaining a deep understanding of crime patterns within specific areas. This research paper endeavors to harness the power of machine learning algorithms to discern and predict criminal activity patterns in particular regions, thus facilitating the expeditious classification of criminal cases and informed decision-making. Drawing on an extensive dataset spanning the past 18 years, culled from diverse reliable sources, this study places a premium on data preprocessing, acknowledging its pivotal role in the predictive process. Feature selection, null value elimination, and label encoding have been meticulously employed to cleanse and prepare the dataset for analysis. The culmination of this research is the presentation of an efficient machine learning model tailored to predict the occurrence of the next criminal case with enhanced accuracy and efficiency.