Fraud Detection in White-Collar Crime

Contents

1 Introduction

1.1 Motivation and problem statement

1.2 Research Methodology

1.3 Goal and structure of the thesis

2 White-Collar Crime

2.1 Fraud Management

2.1.1 Fraud Prevention

2.1.2 Fraud Detection

2.1.3 Fraud Investigation

2.2 Fraud Triangle

2.2.1 Opportunity

2.2.2 Incentive/Pressure

2.2.3 Rationalization/Attitude

3 Types of White-Collar-Crimes

3.1 Fraud

3.2 Credit Card Fraud

3.3 Healthcare Fraud

3.4 Embezzlement

3.5 Criminal Insolvency Offences

3.6 Corruption

4 Data Mining, Text Mining and Big Data

4.1 Introduction into Big Data

4.1.1 The 3 V’s of Big Data

4.1.2 Data Forms

4.2 Data Mining

4.2.1 Types of Machine Learning

4.2.2 Classification of Data Mining Applications

4.3 Text Mining

4.3.1 Practise areas of Text Mining

4.3.2 Example of feature extraction from unstructured data

4.4 Context of Data Mining and Text Mining in White-Collar Crime

5 A case study on Credit Card Fraud Detection

5.1 Overview

5.2 Data Exploration

5.3 Confusion Matrix Terminology

5.4 Algorithms and Techniques

5.4.1 Literature review on Data Mining Techniques

5.4.2 Selection of Data Mining Techniques

5.5 Sampling techniques

5.6 Train and Test Set

5.7 Imbalanced Data

5.7.1 Results on imbalanced Data

5.8 Undersampled Data

5.8.1 Results on undersampled Data

5.9 Oversampled Data

5.9.1 Results on oversampled Data

5.10 Oversampled Data with SMOTE

5.10.1 Results on Oversampled Data with SMOTE

5.11 Undersampled Data with Hyperparameters Optimization

5.11.1 Model Parameter and Hyperparameter

5.11.2 Hyperparameter optimization algorithms

5.11.3 Explanation of selected Hyperparameters

5.11.4 Cross-Validation

5.11.5 Selection of Hyperparameter Optimization Algorithm and k-fold CV

5.11.6 Results on Undersampled Data with Hyperparameter Optimization

5.12 Review of the case study: Credit Card Fraud Detection

6 Conclusion

Research Goal and Focus Areas

The primary research objective of this thesis is to determine which data mining techniques are effective for detecting fraudulent activities in both structured and unstructured datasets, specifically within the context of white-collar crime. The study bridges the gap between theoretical crime analysis and practical technical implementation.

• White-collar crime dynamics and the fraud triangle theory.
• Data mining and text mining applications for fraud detection.
• Big data management and unstructured data transformation.
• Evaluation of machine learning algorithms through a credit card fraud case study.

Excerpt from the Book

Summary of Chapters

1 Introduction: Discusses the growing societal problem of white-collar crime and establishes the research question regarding appropriate data mining techniques for fraud detection.

2 White-Collar Crime: Provides an overview of white-collar crime definitions, management strategies, and introduces the Fraud Triangle by Cressey.

3 Types of White-Collar-Crimes: Examines specific categories such as credit card fraud, embezzlement, insolvency offences, and corruption to illustrate the financial impact of such crimes.

4 Data Mining, Text Mining and Big Data: Explains technical foundations, data formats, and how text mining transforms unstructured data into forms suitable for predictive machine learning models.

5 A case study on Credit Card Fraud Detection: The practical core of the thesis, detailing the application of various data mining algorithms to a real-world credit card dataset using the CRISP-DM model.

6 Conclusion: Summarizes key findings, answers the research question based on the empirical results, and provides recommendations for future research.

Keywords

White-collar crime, Fraud detection, Data mining, Text mining, Big data, Machine learning, CRISP-DM, Credit card fraud, Classification, Unstructured data, Supervised learning, Hyperparameter optimization, Logistic regression, Support vector machine, Neural networks.

Frequently Asked Questions

What is the core focus of this bachelor thesis?

The thesis focuses on using intelligent IT approaches, specifically Data Mining and Text Mining, to identify and mitigate white-collar crimes within large datasets.

Which specific crime types are addressed in the study?

The study examines financial fraud, credit card fraud, healthcare fraud, embezzlement, criminal insolvency, and corruption.

What is the primary research question?

The main question asks which data mining techniques are most appropriate for detecting white-collar crimes in structured and unstructured data.

Which scientific methodology is applied?

The thesis utilizes a literature analysis based on Webster & Watson for the theory, and the CRISP-DM (Cross-Industry Standard Process for Data Mining) reference model for the empirical case study.

What is covered in the main body of the work?

The main body covers the theoretical background of fraud, an introduction to big data and text mining techniques, and a detailed case study on credit card fraud detection.

What are the key technical concepts described in the work?

Key concepts include supervised and unsupervised learning, sampling techniques for imbalanced datasets (e.g., SMOTE), and hyperparameter optimization to enhance predictive model accuracy.

Why is the CRISP-DM model relevant to this research?

It provides a structured, iterative framework (Business Understanding, Data Understanding, Data Preparation, Modelling, Evaluation, and Deployment) for conducting the credit card fraud detection project.

How does text mining contribute to the fraud detection process?

Text mining is essential for converting unstructured data—such as emails or documents—into a structured format (e.g., using Vector Space Model or TF-IDF) so that predictive machine learning algorithms can process them.