Data Mining For Spatio-Temporal Rules With Geospatial Analysis of Crime Patterns

This project establishes a framework based on clustering and association rule mining to detect and analyze trends from temporal and spatial crime activity data. In this model, an open-source GIS application, QGIS, has been used to reveal crime hotspots, and a time series analysis has also been performed to analyze the changing patterns of crime with time.

Code     Project Report

Problem Statement

Increased population, technological advancements and heightened competition for economic resources have created various social problems. Many of these changes in the human condition have brought new challenges to the doorstep of the law enforcement profession that begs for resolution. The major challenge facing law enforcement agencies is to deal with the increased number of criminal activities effectively and efficiently. Current policing strategies work towards finding the criminals, basically after the crime has occurred. However, with the help of technological advancement, we can use historical crime data to recognise crime patterns. If enforcement agencies have a prior assumption of the class of the crime, it would give them tactical advantages and help resolve cases faster. An overall study of criminal activity in a geographic area also helps in understanding the underlying pattern of the crime in that area

Project Objectives

The primary objective of this work is to analyse criminal data based on demographics, spatial and temporal information and consequently identify useful crime patterns to aid police in preventing crimes. Towards this end, this project uses data mining and crime mapping techniques. The main objectives of this project are summarised as follows:

  1. Identifying the crime patterns based on a criminal dataset that contains the geographical location and basic details of the criminal activity.
  2. Exploring data mining techniques to generate association rules for crime analysis.
  3. Visualising these patterns on an open source GIS software - QGIS for better understanding of the results.