MACHINE LEARNING TECHNIQUES BASED EXPLORATION OF VARIOUS TYPES OF CRIMES IN INDIA

Abstract

The reports of crime has been on a rise especially the cyber crime cases have seen a swift increase since the year 2018. It has been reported that there were two lakh incidents in 2018 which almost became seven times by 2021 i.e 14,02,809 cases in 2021; and 2,12,485 incidents in the starting two months of 2022 as per a renowned newspaper. The pace at which crime rate in digital world is increasing has become a pressing issue. As a result, it is critical to employ various techniques to forecast the rate and timing of digital crime events. Hence, in this paper, data for 2018 and 2019 have been collected from six different datasets to research multiple crimes occurring under cyber crime category. The data has been pre-processed and graphically visualized to assess the crime data appropriately. Using R2 and mean square error metrics, six machine learning algorithms have been used to evaluate the performance of the system and it has been discovered that decision trees had the highest R2 value and the lowest mean square error value of 99.9 and 0.01 for all crimes, respectively. In addition, a report on cognizable crime has also been provided for the years 2018 to 2021.