Crime detection using k-means and facial recognition

1Dr. Pradeep Mohan Kumar, S Athul Sriram, Arnold John Regis

176 Views
43 Downloads
Abstract:

The influx of massive amount of data being readily available in the modern world enables us to provide an alternate method of crime detection using data mining. Data mining is the practice of examining preexisting databases in order to generate new information. Many large corporations and businesses are implementing data mining algorithms in view of detecting potential intrusion, fraud or even crime. Even in the year 2020, many organizations still identify intruders by their outer look or by other sensitive attributes like gender, race or religion. Making decisions for criminal activity based on such sensitive attributes will not help to actually detect possible criminals and in fact encourages various kinds of discrimination. A possible alternate system could detect the objective misbehavior of a potential criminal, rather than using irrelevant information like gender, race or religion. Legal data stored in crime databases will be used instead of sensitive attributes. Legal data includes behavior of past crimes done by the individual. If the system identifies any suspected intruder, system will mine the data in database. If the person is detected in the database, their data record will be examined to determine whether it is a harmless person or a potential criminal.

Keywords:

crime, data mining, recidivism, k-means, clustering, facial.

Paper Details
Month3
Year2020
Volume24
IssueIssue 6
Pages4288-4293