Zum Inhalt
Erweiterte Suche
Anmelden
Nach oben

Financial Fraud Detection Using Machine Learning

  • 2025
  • Buch
Verfasst von
Xiyuan Ma
Desheng Wu
Buchreihe
AI for Risks
Verlag
Springer Nature Singapore
Enthalten in
Springer Professional "Wirtschaft+Technik"
Springer Professional "Technik"
Springer Professional "Wirtschaft"
insite
SUCHEN

Über dieses Buch

Dieses Buch dient als umfassender Leitfaden, um verschiedene Aspekte des Finanzbetrugs zu erlernen, einschließlich der damit verbundenen Forschung, der aktuellen Situation, potenzieller Ursachen, des Umsetzungsprozesses, der Erkennungsmethoden, regulatorischer Sanktionen und Managementherausforderungen in börsennotierten Unternehmen. In diesem Buch erfahren die Leser mehr über die betrügerischen Praktiken, die in Unternehmen vorkommen können, die Ausführungsmechanismen, ein Rahmenwerk zur Identifizierung von Indikatoren und verschiedene Erkennungsmethoden, einschließlich qualitativer und quantitativer Modelle. Quantitative Modelle umfassen Diskriminanzanalyse, ökonometrische Analyse und Modelle des maschinellen Lernens (ML). Dieses Buch beleuchtet die Anwendung von ML-Algorithmen zur Erkennung von Finanzbetrug und diskutiert ihre Beschränkungen, wie hohe falsch-positive Kosten, verzögerte Erkennung, die Forderung nach interdisziplinärer Expertise, die Abhängigkeit von spezifischen Anwendungsszenarien und Probleme mit der Qualität von Betrugsdaten. Jedes Kapitel bietet einen strukturierten Überblick über die angesprochenen Probleme, die verwendeten Algorithmen, die experimentellen Ergebnisse und Vergleiche. Darüber hinaus untersucht dieses Buch die Kosten-Nutzen-Abwägungen, mit denen Unternehmen konfrontiert sind, die in Finanzbetrug verwickelt sind, und berücksichtigt Faktoren wie ethische Dilemmata, Chancen, praktische Bedürfnisse, Exposure Risks und Prozesskosten. Dieses Buch richtet sich an Finanzregulierungsinstitutionen, Unternehmensführer, Wirtschaftsprüfer, Wissenschaftler und alle, die an der Aufdeckung von Finanzbetrug interessiert sind. Es bietet praktische Einblicke in die effektive Verhinderung und Kontrolle von Finanzbetrug und einen Überblick über die neuesten Fortschritte bei ML-Technologien. Anhand von Fallstudien aus der realen Welt werden die Leser ein tieferes Verständnis des Finanzbetrugs gewinnen, wie ML verwendet werden kann, um ihn aufzudecken, sowie seiner Fallstricke und Grenzen. Insgesamt schlägt dieses Buch eine Brücke zwischen Theorie und Anwendung und versetzt die Leser in die Lage zu verstehen, wie man Finanzbetrug mit der Macht der Buchhaltung und ML im modernen Geschäftsumfeld aufdecken kann.
Mit KI übersetzt

Inhaltsverzeichnis

Frontmatter
Chapter 1. Introduction
Abstract
This chapter introduces the research background, significance, and the research content of this book. Based on a detailed examination of financial fraud in Chinese listed companies, this chapter systematically analyzes fundamental characteristics of financial fraud, including temporal variations in fraudulent enterprises, severity levels, and industry distribution patterns. This chapter delineates the whole research framework and key innovations, serving as a foundation for subsequent detailed analyses.
Xiyuan Ma, Desheng Wu
Chapter 2. Financial Fraud Research Analysis Based on Bibliometrics and Knowledge Mapping
Abstract
Financial fraud research has grown rapidly over two decades, spanning multiple sectors. Using CiteSpace and VOSviewer, we analyzed thirty years of literature from core journals, identifying key contributors, hotspots, frontiers, and trends to finally conclude some valuable rules and patterns in this kind of research. First, publications show an upward trend, with China and the United States dominating the field. Second, research focuses on three main dimensions including audit practices, legal regulations, and corporate governance, examining fraud motivations, legal consequences, economic impacts, and litigation risks. Third, current frontiers utilize AI technologies like deep learning and language models to build multi-source fraud datasets, improving detection accuracy. These findings establish a foundation for theoretical development, legal frameworks, and advanced detection methods.
Xiyuan Ma, Desheng Wu
Chapter 3. The Definition of Financial Fraud
Abstract
This chapter introduces the definition of financial fraud and financial distress, together with their measurement indicators commonly used in academic papers. It further examines the relationship between distress and fraud, which is a theoretical foundation for subsequent chapters. The definition of financial fraud stems from legal frameworks, emphasizing material misstatements, intentional deception, and resultant economic losses, with varying judicial interpretations across jurisdictions. This book examines three hierarchical categories of financial fraud in Chinese listed companies: CSRC-penalized violations, material misrepresentations with substantial fines, and media-disclosed fraudulent activities. These three categories form the foundational financial fraud datasets for our empirical analyses.
Xiyuan Ma, Desheng Wu
Chapter 4. The Basic Theory of Financial Fraud
Abstract
This chapter reviews the evolution of theories of financial fraud motivation, together with fundamental accounting and management theories such as the double-entry bookkeeping and financial reporting, and the theory of accounting information systems, to systematically explore the driving force of financial fraud and extensively provide theoretical guidance for the establishment of an optimal financial fraud detection indicator system with multi-source information.
Xiyuan Ma, Desheng Wu
Chapter 5. Financial Fraud Litigation and Forensic Accounting
Abstract
This chapter provides a comparative analysis of financial fraud litigation and enforcement frameworks in China and the United States, highlighting their distinctive characteristics. While both systems have evolved through concurrent development of private and public enforcement mechanisms, they exhibit notable differences in their approaches to financial fraud detection and prosecution. Additionally, this chapter elucidates the fundamental principle of forensic accounting to synthesize accounting with legal procedures, which provides a robust foundation for fraud investigation and prosecution.
Xiyuan Ma, Desheng Wu
Chapter 6. Resampling Techniques and Feature Selection
Abstract
This chapter reviews prevalent data preprocessing and feature selection methodologies employed in financial fraud detection research, addressing both class imbalance issues and the ‘curse of dimensionality.‘ Empirical research necessitates comprehensive consideration of data sampling methods, feature selection techniques, and machine learning algorithm selection to balance the multifaceted nature of financial fraud determinants, diversity of feature selection approaches, and scarcity of fraudulent instances.
Xiyuan Ma, Desheng Wu
Chapter 7. Detection Models and Applications
Abstract
This chapter briefly introduces the evolution of financial fraud detection methods from traditional statistical regression through conventional machine learning algorithms to ensemble learning and deep learning algorithms, focusing on mathematical principles, implementation coding and applications comparison.
Xiyuan Ma, Desheng Wu
Chapter 8. Financial Fraud Detection Based on Litigation and Resampling Methods
Abstract
This chapter establishes a multi-modal analytical framework that synthesizes external litigation information, traditional financial reporting, and internal management data, employing regulatory enforcement actions that demonstrate significant correlation with financial fraud as proxy variables to construct an imbalanced large-scale dataset.This chapter explores financial fraud detection models by integrating resampling techniques with machine learning methods, considering scenarios with and without feature selection, to identify optimal combinations of resampling methods and classification models. Furthermore, the role of litigation factors in financial fraud detection within immature legal environments is examined, providing a new research perspective for financial fraud detection studies.
Xiyuan Ma, Desheng Wu
Chapter 9. Financial Fraud Detection Based on Feature Selection and the GONE Framework
Abstract
This chapter sets strict standard for financial fraud to only include major accounting fraud including false records, misleading statements and material omissions which is identified based on violation typologies and penalty magnitude into the analytical framework. Leveraging the theory of GONE (Greed, Opportunity, Need, Exposure), this chapter implements an innovative two-stage feature selection methodology integrating Pearson correlation coefficients and average feature importance scores, simultaneously validating theoretical foundations of the GONE theory and optimizing machine learning model performance.
Xiyuan Ma, Desheng Wu
Chapter 10. The Classical Case of Financial Fraud
Abstract
This chapter presents an in-depth analysis of seven landmark financial fraud cases involving companies such as LeTV and Luckin Coffee. Each case study provides comprehensive details of the fraudulent methodologies employed and the subsequent regulatory actions. These cases offer readers profound insights into the distinctive characteristics of financial fraud in China, notably its prolonged duration, extensive falsification of records, and the involvement of multiple stakeholders. Specifically, background information on each company the fraudulent activities identified by regulatory authorities, alongside the corresponding penalties imposed at both corporate and individual levels are presented, enabling readers to quickly grasp the key points of each financial fraud scandal. This chapter concludes by proposing context-specific recommendations for preventing similar financial fraud incidents, with particular attention to unique market environment and institutional framework in China.
Xiyuan Ma, Desheng Wu
Titel
Financial Fraud Detection Using Machine Learning
Verfasst von
Xiyuan Ma
Desheng Wu
Copyright-Jahr
2025
Verlag
Springer Nature Singapore
Electronic ISBN
978-981-9508-40-2
Print ISBN
978-981-9508-39-6
DOI
https://doi.org/10.1007/978-981-95-0840-2

Die PDF-Dateien dieses Buches wurden gemäß dem PDF/UA-1-Standard erstellt, um die Barrierefreiheit zu verbessern. Dazu gehören Bildschirmlesegeräte, beschriebene nicht-textuelle Inhalte (Bilder, Grafiken), Lesezeichen für eine einfache Navigation, tastaturfreundliche Links und Formulare sowie durchsuchbarer und auswählbarer Text. Wir sind uns der Bedeutung von Barrierefreiheit bewusst und freuen uns über Anfragen zur Barrierefreiheit unserer Produkte. Bei Fragen oder Bedarf an Barrierefreiheit kontaktieren Sie uns bitte unter accessibilitysupport@springernature.com.