Prediction of Crime Data using Machine Learning Techniques

Crime has a devastating effect on society and can take many forms, ranging from violent acts such as murder and assault to less serious but still damaging offenses like burglary and vandalism. Crime records are records of various criminal activities in a given area. These records typically include information such as the date, time, and location of reported crime, the nature of the crime, and the name of the suspect. However, there are many problems associated with keeping accurate and up-to-date records of criminal activity. The most common issue is a lack of uniformity in the way different states and municipalities track and record criminal activities. Additionally, many states and municipalities lack the resources to properly maintain accurate records, resulting in incomplete or inaccurate information. Finally, maintaining crime records is time consuming and expensive, leading to a backlog in the processing of reported crimes. This can complicate investigations and create a backlog of cases that need to be addressed. Therefore, this study proposes a model with an objective to make criminal investigation more effective. The two main facets of this research are crime analysis and prediction. Based on these aspects the datasets which contain all the data is pre-processed using different pre-processing techniques. Further, the dataset is processed using machine learning algorithms like Support Vector Machine, Random forest classifier, Decision tree, and K-Means. In conclusion, crime data analysis and prediction can provide valuable insights into criminal activity and aid in the development of effective crime prevention strategies.