Crime analysis in India using data mining techniques

Abstract

An approach for crime detection in India using Data mining techniques is proposed in this paper. The approach consists of the following steps -Data pre-processing, clustering, classification and visualization. Data mining techniques are often applied to Criminology as it provides good results. Criminology is a field which studies about various crime characteristics. Analyzing crime data means exploring crime data. Crime is identified using k-means clustering and the clusters are formed based on the similarity of the crime attributes. The Random Forest algorithm and Neural networks are applied on the data for classification. Visualization is achieved using the Google marker clustering and the crime spots are marked on the India map. The accuracy is verified using WEKA tool. This approach will benefit the Crime department of India in analyzing crime with better prediction. The paper focuses on the crime analysis of various Indian states and union territories during 2001 to 2012.

A Review on Resolving Crime with Prediction

“…The paper focuses on crime analysis between 2001 and 2012 in different Indian and the Union States. [16] Charlie Catlett a,b , Eugenio Cesario c,d,* , Domenico Talia e , Andrea Vinci d : This document provides a prediction-based approach to identify high-risk criminality regions in urban areas automatically and accurately forecast patterns in each field, based on spatial analysis and automated model regressive. The algorithm result is a model for the projection of spatial-temporal crime consisting of a set of crime-dense areas with associated crime predictors, each of which is a predictive model for estimating the number of crimes in their associated field.…”