Title: Identification of authorship and prevention of fraudulent transactions/cybercrime using efficient high performance machine learning techniques

Authors: B.J. Sowmya; R. Hanumantharaju; D. Pradeep Kumar; K.G. Srinivasa

Addresses: Department of Computer Science and Engineering, M.S. Ramaiah Institute of Technology, India; Affiliated to: Visvesvaraya Technological University, India ' Department of Computer Science and Engineering, M.S. Ramaiah Institute of Technology, India; Affiliated to: Visvesvaraya Technological University, India ' Department of Computer Science and Engineering, M.S. Ramaiah Institute of Technology, India; Affiliated to: Visvesvaraya Technological University, India ' Department of Information Management and Emerging Engineering, National Institute of Technical Teachers Training and Research, India

Abstract: Cyber safety is the best skill required among a group of employees in organisations. Many offenders hide behind anonymous masking, including dishonest purchases, brazen plagiarism of the work, breaching corporate safety and stealing private data. To understand and analyse the actual phenomenon encountered with data, requirements of scientific methods, machine learning techniques, processes are to be used. This paper aims to tackle these problems by providing a protection layer for users, where data is being gathered from cyber security sources, analytical complement with latest data-driven patterns provides effective security solutions. We then discuss machine learning, deep learning powered models for the detection of insider threats and identifying authorship identification of anonymised articles. The individual modules are trained on authorship attribution, mouse monitoring, keyboard monitoring and command tracing and reached promising results with good accuracies in the range of 65%-85% on average.

Keywords: cybercrime; authorship attribution; machine learning; authorship identification; deep learning.

DOI: 10.1504/IJBIDM.2023.127312

International Journal of Business Intelligence and Data Mining, 2023 Vol.22 No.1/2, pp.144 - 169

Received: 29 Aug 2021
Accepted: 13 Dec 2021
Published online: 30 Nov 2022 *