Machine Learning-Based Prediction Analysis of Unlawful Activities to Aid Law Enforcement

Machine Learning-Based Prediction Analysis of Unlawful Activities to Aid Law Enforcement

Vijayalakshmi G. V. Mahesh, Shilpa Hiremath, Chandra Prabha R.
Copyright: © 2024 |Pages: 18
DOI: 10.4018/979-8-3693-0220-0.ch011
OnDemand:
(Individual Chapters)
Available
$33.75
List Price: $37.50
Current Special Offers
10% Discount:-$3.75
TOTAL SAVINGS: $3.75

Abstract

One of our society's most significant and pervasive issues is crime. Numerous crimes are perpetrated often each day. The development of policing strategies and the implementation of crime prevention and control depend greatly on crime prediction. The most popular prediction technique right now is machine learning. Little research, however, has rigorously contrasted various machine learning approaches for crime prediction. The dataset in this instance consists of the date and the annual crime rate for the corresponding years. The crime rate used in this project is only based on robberies. Utilising historical data, the authors employ the linear and random forest regression algorithms to estimate future crime rates. The algorithm receives the date as input, and the result is the total number of crimes that year.
Chapter Preview
Top

To address the issue of crime prediction and to minimize crime, various researchers have offered numerous approaches, and several crime-prediction algorithms have also been proposed. The type of data used and the attributes used for prediction affect the prediction's accuracy.

According to (Bogomolov et al, 2014), mobile network activity was utilized to gather information on human behavior that was then used to predict the hotspot for crime in London with an accuracy of roughly 70%. The key innovation of the suggested method was how it approached the problem of crime prediction by using aggregated and anonymized human behavioral data obtained from mobile network activity. The results of earlier research projects, which either used historical context or offender profiling, reinforced the idea that aggregated human behavioral data collected from the mobile network infrastructure.

The first study to use data mining to analyze and forecast crime was by (Sathyadevan et al, 2014). Here the Naive Bayes algorithm and decision trees were used to predict and categorize crime using data gathered from various websites and newsletters, and it was discovered that Naive Bayes algorithm performed better. The algorithm could identify places with a high likelihood of crime and could depict crime-prone zones.

Complete Chapter List

Search this Book:
Reset