Use of Artificial Intelligence to support cybercrime research
Use of Artificial Intelligence to support cybercrime research
Efficient processing of the ever-growing volume of cybercrime related posts, images and videos appearing online is a significant challenge. This chapter discusses the trend for criminology researchers to use artificial intelligence (AI) algorithms for relevance pre-filtering and quantitative analysis of content at scales larger than can be achieved working manually. Algorithms can be used as black box AI tools, or integrated into socio-technical AI methodologies where manual criminology analysis is used for content and observed behavior patterns that are hard for machines to understand. AI must be deployed with care however, balancing the advantages in terms of scaling up the criminology analysis with the possibility of introducing new sources of algorithmic bias and error. Two concrete case studies are presented describing socio-technical AI approaches.
Artificial intelligence, Machine learning, Natural Language Processing, Information Extraction, Socio-technical, facial recognition software, Object detection, Event Detection, Social Network Analysis, Predictive Policing
213-232
Middleton, Stuart
404b62ba-d77e-476b-9775-32645b04473f
1 July 2021
Middleton, Stuart
404b62ba-d77e-476b-9775-32645b04473f
Middleton, Stuart
(2021)
Use of Artificial Intelligence to support cybercrime research.
In,
Lavorgna, Anita and Holt, Thomas
(eds.)
Researching Cybercrimes: Methodologies, Ethics, and Critical Approaches.
1 ed.
Palgrave Macmillan, .
(doi:10.1007/978-3-030-74837-1).
Record type:
Book Section
Abstract
Efficient processing of the ever-growing volume of cybercrime related posts, images and videos appearing online is a significant challenge. This chapter discusses the trend for criminology researchers to use artificial intelligence (AI) algorithms for relevance pre-filtering and quantitative analysis of content at scales larger than can be achieved working manually. Algorithms can be used as black box AI tools, or integrated into socio-technical AI methodologies where manual criminology analysis is used for content and observed behavior patterns that are hard for machines to understand. AI must be deployed with care however, balancing the advantages in terms of scaling up the criminology analysis with the possibility of introducing new sources of algorithmic bias and error. Two concrete case studies are presented describing socio-technical AI approaches.
This record has no associated files available for download.
More information
Published date: 1 July 2021
Keywords:
Artificial intelligence, Machine learning, Natural Language Processing, Information Extraction, Socio-technical, facial recognition software, Object detection, Event Detection, Social Network Analysis, Predictive Policing
Identifiers
Local EPrints ID: 452593
URI: http://eprints.soton.ac.uk/id/eprint/452593
PURE UUID: a80a801d-4c99-42c8-b78a-1e87b56c8a8f
Catalogue record
Date deposited: 11 Dec 2021 11:28
Last modified: 12 Apr 2024 01:37
Export record
Altmetrics
Contributors
Editor:
Anita Lavorgna
Editor:
Thomas Holt
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
