Analysing Crimes of Indian Datasets Based on Machine Learning Methods

Abstract

Present scenario problems were facing the crime rate increases day by day, due to crimes being a typical social issue influencing personal satisfaction and the economic development of the general public. It is a fundamental factor that decides if individuals move to another city and what spots should be kept away from when they travel. With the expansion of violations, law requirement organizations are proceeding to request progressed geographic data frameworks and new information mining to deal with enhanced crime examination and better secure their groups.  Even though violations could happen all over, offenders must deal with wrongdoing openings they look in most commonplace territories for them. Therefore, it becomes vital to study the factors that impact the crime rate. To be better prepared to respond to criminal activity, it is important to understand the patterns in crime; this study analyzes crime data from all over the nation, available publicly on websites such as data.gov.in or Kaggle. This study investigates the relationship between various factors and the crime rate in India and focuses on the extent of effects of various factors on crime registered under IPC or SLL in all Indian states and the union territories. The study covers the data of all the Indian states and the union territories of periods ranging from 2001 to 2014. However, different classes of crimes have a slightly different range of years. The data for district-wise crimes in India is from 2001 to 2014, the data for crimes against SC, ST & Children is from 2001 to 2012, and the data for murder victims by age and gender is from 2001 to 2010. The findings show that these factors are crucial determinants of the rate of criminal cases registered in India. We unequivocally believe that discovering connections between wrongdoing components could profoundly help foresee the potentially risky hotspots later on.