White-collar crime is and has always been an urgent issue for the society. In recent years, white-collar crime has increased dramatically by technological advances. The studies show that companies are affected annually by corruption, balance-sheet manipulation, embezzlement, criminal insolvency and other economic crimes. The companies are usually unable to identify the damage caused by fraudulent activities. To prevent fraud, companies have the opportunity to use intelligent IT approaches. The data analyst or the investigator can use the data which is stored digitally in today’s world to detect fraud.
In the age of Big Data, digital information is increasing enormously. Storage is cheap today and no longer a limited medium. The estimates assume that today up to 80 percent of all operational information is stored in the form of unstructured text documents. This bachelor thesis examines Data Mining and Text Mining as intelligent IT approaches for fraud detection in white-collar crime. Text Mining is related to Data Mining. For a differentiation, the source of the information and the structure is important. Text Mining is mainly concerned with weak- or unstructured data, while Data Mining often relies on structured sources.
At the beginning of this bachelor thesis, an insight is first given on white-collar crime. For this purpose, the three essential tasks of a fraud management are discussed. Based on the fraud triangle of Cressey it is showed which conditions need to come together so that an offender commits a fraudulent act. Following, some well-known types of white-collar crime are considered in more detail.
Text Mining approach was used to demonstrate how to extract potentially useful knowledge from unstructured text. For this purpose, two self-generated e-mails were converted into struc-tured format. Moreover, a case study will be conducted on fraud detection in credit card da-taset. The dataset contains legitimate and fraudulent transactions. Based on a literature research, Data Mining techniques are selected and then applied on the dataset by using various sampling techniques and hyperparameter optimization with the goal to identify correctly pre-dicted fraudulent transactions. The CRISP-DM reference model was used as a methodical procedure.
Inhaltsverzeichnis (Table of Contents)
- Management Summary
- Introduction
- Motivation and problem statement
- Research Methodology
- Goal and structure of the thesis
- White-Collar Crime
- Fraud Management
- Fraud Prevention
- Fraud Detection
- Fraud Investigation
- Fraud Triangle
- Opportunity
- Incentive/Pressure
- Rationalization/Attitude
- Types of White-Collar-Crimes
- Fraud
- Credit Card Fraud
- Healthcare Fraud
- Embezzlement
- Criminal Insolvency Offences
- Corruption
- Data Mining, Text Mining and Big Data
- Introduction into Big Data
- The 3 V's of Big Data
- Data Forms
- Data Mining
- Types of Machine Learning
- Classification of Data Mining Applications
- Text Mining
- Practise areas of Text Mining
- Example of feature extraction from unstructured data
- Context of Data Mining and Text Mining in White-Collar Crime
- A case study on Credit Card Fraud Detection
- Overview
- Data Exploration
- Algorithms and Techniques
- Literature review on Data Mining Techniques
- Selection of Data Mining Techniques
- Sampling techniques
- Train and Test Set
- Imbalanced Data
- Results on imbalanced Data
- Undersampled Data
- Results on undersampled Data
- Oversampled Data
- Results on oversampled Data
- Oversampled Data with SMOTE
- Results on Oversampled Data with SMOTE
- Undersampled Data with Hyperparameters Optimization
- Model Parameter and Hyperparameter
- Hyperparameter optimization algorithms
- Explanation of selected Hyperparameters
- Cross-Validation
- Selection of Hyperparameter Optimization Algorithm and k-fold CV
- Results on Undersampled Data with Hyperparameter Optimization
- Review of the case study: Credit Card Fraud Detection
Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)
This thesis examines the application of Data Mining and Text Mining techniques to detect fraud in white-collar crime. The work aims to demonstrate the potential of these intelligent IT approaches in combating economic offenses. It explores how these techniques can be used to extract valuable insights from both structured and unstructured data, enabling the identification of fraudulent activities.
- The use of Data Mining and Text Mining in fraud detection
- The impact of technological advancements on white-collar crime
- The application of these techniques to real-world datasets, specifically credit card fraud
- The importance of data exploration and pre-processing in fraud detection
- The effectiveness of various Data Mining algorithms in identifying fraudulent transactions
Zusammenfassung der Kapitel (Chapter Summaries)
- Introduction: This chapter introduces the motivation behind the research and outlines the research methodology. It defines the scope of the thesis and sets forth the objectives and structure.
- White-Collar Crime: This chapter provides an overview of white-collar crime, specifically focusing on fraud management. It discusses the essential tasks of fraud prevention, detection, and investigation. It also explores the Fraud Triangle, outlining the conditions that contribute to fraudulent acts.
- Types of White-Collar-Crimes: This chapter delves into various types of white-collar crimes, including fraud, credit card fraud, healthcare fraud, embezzlement, criminal insolvency offenses, and corruption. It provides a brief description of each type of crime and its significance.
- Data Mining, Text Mining and Big Data: This chapter introduces the concepts of Big Data, Data Mining, and Text Mining. It describes the characteristics of Big Data and explores the types of machine learning used in Data Mining. It further explains the practice areas of Text Mining and illustrates how feature extraction can be performed on unstructured data.
- A case study on Credit Card Fraud Detection: This chapter presents a case study on credit card fraud detection. It outlines the data exploration process, the algorithms and techniques used, and the different sampling methods employed. The chapter explores the results of the case study using various evaluation metrics and discusses the challenges of dealing with imbalanced datasets.
Schlüsselwörter (Keywords)
The main focus of this thesis lies in the intersection of white-collar crime, fraud detection, Data Mining, Text Mining, and Big Data. The study explores the use of intelligent IT approaches to identify and prevent economic offenses by analyzing both structured and unstructured data. Key concepts include credit card fraud, fraud management, machine learning algorithms, sampling techniques, and hyperparameter optimization.
- Citation du texte
- Rohan Ahmed (Auteur), 2017, Fraud Detection in White-Collar Crime, Munich, GRIN Verlag, https://www.grin.com/document/426831
