Survey on Crime Analysis and Prediction Using Data Mining and Machine Learning Techniques

Abstract

Crime is an unlawful event which affects the harmony of humanity. Whoever got victimized in a crime, it affects them both physically and mentally. Hence, they are haunted by the memories throughout their life. Due to the limitations, traditional data collection and analysis methods are not very effective now. Yesteryears, the researches were concentrating on demographic features of the population. Nowadays, the dynamic characteristics of individual or specific group could easily be extracted from the search engines, social media, e-commerce platforms, mobile applications, IOT devices, surveillance cameras, sensors and geographical information systems. The recent technological advancements are helpful in integration of data from various sources, classification of information into granular level, identification of crime sequences and designing a framework. Particularly, the artificial intelligence methodology called deep learning imitates the functions of human brain and able to acquire knowledge from unstructured data. It makes revolutionary changes in crime forecasting, predictive policing and legal strategy formulations. The following survey explores the possibilities of scrutinizing the data from huge repositories, analyzing various socioeconomic factors associated to the crime incidents, identifying the outliers, categorizing the patterns and designing effective computational models to predict crimes by using data mining and machine learning techniques.