Analysis of criminal spatial events in india using exploratory data analysis and regression

Abstract

Crime against women (CAW) in India is the violence against women that is at par in previous years. India is densely populated has added to the figures of crime against women. This paper aims at study of crime against women dataset given by NCRB (National Crime Record Bureau) from 2001 to 2020 for all the 27 states and 9 union territories. EDA (exploratory data analysis) with linear regression is a powerful combination for understanding the relationship between various factors and the incidence of crime against women. EDA is a process of analysing and summarizing the main characteristics of a data set through visualizations, descriptive statistics, and other techniques. At the same time, linear regression is a statistical method that models the relationship between a dependent variable and one or more independent variables. India's crime against women dataset on various crime categories under Indian Penal Code (IPC) such as rape, cruelty by husband and his relatives, kidnapping and abduction, dowry deaths, assault on women with intent to outrage her modesty, insult to modesty of women and human trafficking are considered to accomplish this. CRISP-DM methodology allows for a consistent and structured approach to data mining, which reduces the risk of errors and improves the chances of success in predicting crime rate. The proposed model has various data analytics steps to pre-process the datasets and visualize the crime rate. The visualization of data helps to uncover trends present in the crime dataset. The proposed predictive model analyses data and predict crime against women under four IPC categories to give accuracy of 72.29, 92.15, 83.30 and 84.33% respectively.