Here is the article rewritten in markdown format:

Financial Fraud Detection using Machine Learning Techniques

Background

The prevalence of financial fraud has led to significant losses for individuals and organizations alike. Types of financial fraud include credit card fraud, securities and commodities fraud, mortgage fraud, corporate fraud, money laundering, and cryptocurrency fraud. To mitigate these losses, new methods are needed to prevent further attacks.

Research Question 2 (RQ2)

Machine Learning Methods for Financial Fraud Detection

The following machine learning techniques have been employed in literature for financial fraud detection:

• Support Vector Machine (SVM):
  • A supervised machine learning method that seeks to find the maximum margin hyperplane for classifying input training data into two categories.
• Fuzzy-Logic-Based Method:
  • An effective conceptual framework for addressing uncertainty and ambiguity in data representation.
• Hidden Markov Model (HMM):
  • A dual embedded random method used to perform complex random processes better than traditional Markov models.

Studies using these methods

The following studies have employed the above machine learning techniques for financial fraud detection:

• Rajak and Mathai’s hybrid technique based on SVM and the fusion Danger theory
• Francis et al.’s automated medical bill architecture using SVM
• Xu and Liu’s optimized SVM for detecting fraudulent activities in online credit card transactions
• Mareeswari and Gunasekaran’s approach combining SVM and spike detections for detecting fraudulent behaviors in credit cards
• Sundarkumar et al.’s one-class SVM-based under-sampling technique for enhancing fraud detection in insurance industries

Let me know if you have any further requests!