Design of a financial fraud detection model optimized by multi-task learning and graph neural networks


Abstract

Contemporary financial regulation and risk identification are increasingly challenged by the escalating intricacy of inter-firm relational architectures, the diversification of financial conduct, and the multidimensionality of data sources. Conventional fraud detection methodologies, predominantly grounded in single-task paradigms and static heuristic indicators, are insufficient to holistically capture the complex manifestations of fraudulent corporate behavior, encompassing both financial anomalies and behavioral aberrations. To surmount these limitations, this study introduces GN-MTNet, a novel financial fraud detection framework that synthesizes graph neural networks with multi-task learning. The proposed architecture constructs enterprise relational graphs, employs graph-based neural encoders to extract high-order structural representations, and concurrently addresses three core tasks: fraud identification, anomaly quantification, and behavioral deviation classification. A unified, shared multi-task learning framework is devised to encapsulate firm-level irregularities from diverse analytical perspectives, thereby facilitating the synergistic optimization of risk detection and pattern discernment. Empirical evaluations conducted on two benchmark datasets—the Financial Statement Fraud Dataset and OpenCorporates+AMiner—demonstrate that GN-MTNet markedly surpasses existing approaches in terms of classification precision, anomaly reconstruction capability, and multi-task synergy. Ablation studies further substantiate the critical contributions of graph-based modeling, the task-sharing mechanism, and the composite loss formulation to the model’s holistic efficacy. Collectively, this methodology offers a more nuanced and intelligent paradigm for financial fraud detection, furnishing essential technical underpinnings for the development of enterprise risk profiling and the enhancement of intelligent financial auditing systems.
Ask to review this manuscript

Notes for potential reviewers

  • Volunteering is not a guarantee that you will be asked to review. There are many reasons: reviewers must be qualified, there should be no conflicts of interest, a minimum of two reviewers have already accepted an invitation, etc.
  • This is NOT OPEN peer review. The review is single-blind, and all recommendations are sent privately to the Academic Editor handling the manuscript. All reviews are published and reviewers can choose to sign their reviews.
  • What happens after volunteering? It may be a few days before you receive an invitation to review with further instructions. You will need to accept the invitation to then become an official referee for the manuscript. If you do not receive an invitation it is for one of many possible reasons as noted above.

  • PeerJ Computer Science does not judge submissions based on subjective measures such as novelty, impact or degree of advance. Effectively, reviewers are asked to comment on whether or not the submission is scientifically and technically sound and therefore deserves to join the scientific literature. Our Peer Review criteria can be found on the "Editorial Criteria" page - reviewers are specifically asked to comment on 3 broad areas: "Basic Reporting", "Experimental Design" and "Validity of the Findings".
  • Reviewers are expected to comment in a timely, professional, and constructive manner.
  • Until the article is published, reviewers must regard all information relating to the submission as strictly confidential.
  • When submitting a review, reviewers are given the option to "sign" their review (i.e. to associate their name with their comments). Otherwise, all review comments remain anonymous.
  • All reviews of published articles are published. This includes manuscript files, peer review comments, author rebuttals and revised materials.
  • Each time a decision is made by the Academic Editor, each reviewer will receive a copy of the Decision Letter (which will include the comments of all reviewers).

If you have any questions about submitting your review, please email us at [email protected].