A Financial Fraud Detection Model Based on LSTM Deep Learning Technique. Issue 4 (1st October 2020)

• Record Type: Journal Article
• Authors: Alghofaili, Yara; Albattah, Albatul; Rassam, Murad A.
• Is Part Of: Journal of applied security research. Volume 15:Issue 4(2020)
• Page Start: 498
• Page End: 516
• Publication Date: 2020-10-01
• Subjects: Financial fraud -- fraud detection -- deep learning -- long short-term memory; Police, Private -- Training of -- Periodicals; Police, Private -- Training of -- United States -- Periodicals; Private security services -- Periodicals; 363.289
• DOI: 10.1080/19361610.2020.1815491
• Languages: English
• ISSNs: 1936-1610

Abstract

Abstract: As the use of the internet is growing exponentially, more and more businesses such as the financial sector are operationalizing their services online. Consequently, financial frauds are increasing in number and forms around the world, which results in tremendous financial losses which make financial fraud a major problem. Unauthorized access and irregular attacks are examples of threats that should be detected by means of financial fraud detection systems. Machine learning and data mining techniques have been widely used to tackle this issue over the past few years. However, these methods still need to be improved in terms of speed computation, dealing with big data, and identify the unknown attack patterns. Therefore, in this paper, a deep learning-based method is proposed for the detection of financial fraud based on the Long Short-Term Memory (LSTM) technique. This model is aimed at enhancing the current detection techniques as well as enhancing the detection accuracy in the light of big data. To evaluate the proposed model, a real dataset of credit card frauds is utilized and the results are compared with an existing deep learning model named Auto-encoder model and some other machine learning techniques. The experimental results illustrated a prefect performance of LSTM where it achieved 99.95% of accuracy in less than a minute.