Online ISSN- 2457-0818

Vol 5, No 2 (2020)

Random Forest and ARIMA model for Crime Prediction and Analysis

Authors: Anushka Sancheti, Shambhavi Shete, Pranati Thakare, Saloni Goya, Sandhya Arora

Abstract: Crime is one of the biggest and dominating problems of our society and the whole world is struggling to fight against crime. The police department, cyber cell and other crime control agencies face problems in surveillance action for crime prevention. Processing, capturing and analyzing security-related data is increasingly difficult. That's why much of the scattered crime data is not being used. In this work, the technique of machine learning and data science for crime prediction is used and for testing the Toronto crime data set is used. The crime data is extracted from the official portal of Toronto police. Data preprocessing, data modelling is done and many machine learning algorithms are applied for crime prediction. The random forest algorithm and ARIMA model worked on crime data with good accuracy. Results are analyzed and visualized in terms of graphical representation of many cases, for example, at which time the crime rates are high or at which month, area and location the criminal activities are high. The sole purpose of this work is to explain how machine learning can be used by law enforcement agencies to detect, predict, and solve crimes faster and thus reduce the crime rate.

Full Issue

View or download the full issue PDF

Table of Contents