Fraud404: A Fuzzy Approach for Financial Statement Fraud Detection in Resource-Constrained Settings

Organizations, ranging from resource-constrained startups to established institutions, are susceptible to financial statement fraud. Traditional detection methods often lack transparency and struggle with complex fraud schemes. Fraud404, an approach leveraging fuzzy logic, offers a solution applicable across various organizational landscapes. This model analyzes key financial indicators and relationships within statements, effectively capturing subtle patterns and inconsistencies that might signal potential fraud and generate an overall fraudulent degree for the input financial statements. Unlike opaque machine learning methods, Fraud404 prioritizes interpretability, generating clear explanations for its decision-making process and fostering trust within the financial ecosystem. This transparency empowers stakeholders to understand the reasoning behind potential fraud flags, enabling informed decision-making. Additionally, Fraud404 possesses the capability to dynamically calculate the optimal threshold based on the relative importance of precision, recall, and F1-score. This flexibility allows organizations to tailor Fraud404's behavior to their specific risk tolerance and fraud detection priorities. Fraud404 operates efficiently with O(n) time complexity, enabling faster fraud identification. In essence, Fraud404 addresses the limitations of traditional methods by employing fuzzy logic and prioritizing interpretability. This approach, achieving a high F1-score of 0.996 at a threshold of 52.419, offers exceptional performance. At this specific threshold, Fraud404 achieves a precision of 1, indicating that all statements flagged as fraudulent are indeed confirmed frauds, while also maintaining a high recall of 0.992, which indicates that the model successfully identifies a very large proportion of true fraud cases, minimizing the chances of overlooking actual fraudulent activity. By empowering faster and more accurate fraud detection, Fraud404 fosters a more secure and ethical financial environment. This, in turn, strengthens the foundation for organizational success and growth within a robust economic system.

Authors

Institutions

Publication Details

Published
2025-04-13
DOI
https://doi.org/10.1109/iccit63348.2025.10989364
Citations
1
Primary Topic
Imbalanced Data Classification Techniques
Type
article
Field-Weighted Citation Impact
2.48
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Fraud404: A Fuzzy Approach for Financial Statement Fraud Detection in Resource-Constrained Settings

Shorouq Alansari, Sumayyah Albarakati, Shomokh Althagafi, Reyouf Wasel et al.
1 citations
Imbalanced Data Classification Techniques
2.48
article

Fraud404: A Fuzzy Approach for Financial Statement Fraud Detection in Resource-Constrained Settings

Shorouq Alansari, Sumayyah Albarakati, Shomokh Althagafi, Reyouf Wasel, Doja Alhebshi
article en
1 citations

Abstract

Organizations, ranging from resource-constrained startups to established institutions, are susceptible to financial statement fraud. Traditional detection methods often lack transparency and struggle with complex fraud schemes. Fraud404, an approach leveraging fuzzy logic, offers a solution applicable across various organizational landscapes. This model analyzes key financial indicators and relationships within statements, effectively capturing subtle patterns and inconsistencies that might signal potential fraud and generate an overall fraudulent degree for the input financial statements. Unlike opaque machine learning methods, Fraud404 prioritizes interpretability, generating clear explanations for its decision-making process and fostering trust within the financial ecosystem. This transparency empowers stakeholders to understand the reasoning behind potential fraud flags, enabling informed decision-making. Additionally, Fraud404 possesses the capability to dynamically calculate the optimal threshold based on the relative importance of precision, recall, and F1-score. This flexibility allows organizations to tailor Fraud404's behavior to their specific risk tolerance and fraud detection priorities. Fraud404 operates efficiently with O(n) time complexity, enabling faster fraud identification. In essence, Fraud404 addresses the limitations of traditional methods by employing fuzzy logic and prioritizing interpretability. This approach, achieving a high F1-score of 0.996 at a threshold of 52.419, offers exceptional performance. At this specific threshold, Fraud404 achieves a precision of 1, indicating that all statements flagged as fraudulent are indeed confirmed frauds, while also maintaining a high recall of 0.992, which indicates that the model successfully identifies a very large proportion of true fraud cases, minimizing the chances of overlooking actual fraudulent activity. By empowering faster and more accurate fraud detection, Fraud404 fosters a more secure and ethical financial environment. This, in turn, strengthens the foundation for organizational success and growth within a robust economic system.

Umm al-Qura University (SA)
Peace, Justice and strong institutions
Openalex Percentile: Top 10%
Imbalanced Data Classification Techniques
2.48
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.