1.1 Introduction
Fraud is the deliberate deception practiced with a view to gaining an unlawful or unfair advantage. The crave for information technology globally has greatly had an influence in our banking sectors. Thus a fest and quick development has erupted in many countries of the world in terms of automation advancement and technology. Such advancement includes the design and implementation of an automated fraud detection system which is witnessed in our various insurance sectors worldwide. In as much as financial sector is concerned, there should be possible ways made to avoid fraudulent from gaining access to the banking and all other insurance sectors. By the introduction of an automated computer base security system, it will aid to eliminate totally the idea and practice of fraud in banking industries.
As a prelude to other parts of this study, this chapter will discuss the background upon which this study was initiated, the statement of problems that led to this study, the Aim and Objectives of the study. Others are Significance of the study, Scope of work, Research hypothesis and questions, Limitations of the Study and Definition of technical terms.
1.2 Background of Study
Fraud is committed in various fields such as insurance (Ormerod et al. 2010; Li et al. 2008; Atwood et al. 2006), credit card (Weston et al. 2008; Dal Pozzolo et al. 2014), telecommunications (Estevez, 2006), and financial communications (Kirkos et al. 2007; Kotsiantis et al. 2006; Holton, 2009). Insurance fraud is one of the most frequent types of fraud to undertake. This type of fraud can take place in many forms with the simple objective of gaining money (Almedia, 2009). One of these domains is car insurance in which fraudsters (policyholders) setting planned traffic accidents up and file fake insurance claims (e.g. inflating costs) to obtain an illicit benefit from their insurance policy (Ayuso et al. 2011). It has been reported that Almost 21% to 36% of auto-insurance claims contain elements of suspected fraud but only less than 3% of them are prosecuted (Nian et al. 2016).
The high rate of e-payment fraud has called for stronger measures to be applied in detecting fraud. Deep learning is considered to be one of the measures that can be successfully applied for the detecting of e-payment fraud, financial fraud detection and anti-money laundering. Deep learning is a class of machine learning algorithms that use a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input; learn in supervised (e.g., classification) and/or unsupervised (e.g., pattern analysis) manners; learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts; Deng, L.; Yu, D. (2014).
Deep learn leverages both supervised learning techniques, such as the classification of suspicious transactions, and unsupervised learning, e.g. anomaly detection. The study seeks to appraise application of deep learning for fraud detection in e-payment system.
Traditional systems for fraud detection are only able to find fraudulent customers (opportunistic fraud), whereas more professional fraudsters will be overlooked (a white paper, Roberts, 2010). In other words, opportunistic fraud is a continuous issue for insurers, whereas the more remarkable challenge comes from professional fraud, and such organized groups of perpetrators impose the greatest cost upon insurers. Fraudulent groups are being arranged by fraudsters in order to employ different individuals for doing some works, and using the newest technologies to be at least one step in front of insurers. They know properly that insurers and law enforcements officials utilize what kind of tools, and information (Smallwood and Breading, 2011). Due to aforementioned reasons, it is imperative for insurance companies to consider relevant methods for finding organized fraud groups, and promulgating them in the future.
Therefore, in Nigeria where the research was carried out, the activities that was conducted is to know the Application of Deep Learning for Fraud Detection in e-Payment System.
1.3 Statement of Problems
Investigation revealed that the level of fraud emanating from e-payment transaction is at an alarming rate. A recent report shows that Credit card fraud resulted in the loss of $3 billion to North American financial institutions in 2017. The increasing use of digital payments systems such as Apple Pay, Android Pay, and Venmo has result to increases in fraudulent activity.
Deep Learning presents a promising solution to the problem of credit card fraud detection by enabling institutions to make optimal use of their historic customer data as well as real-time transaction details that are recorded at the time of the transaction.
Deep anti-money laundering detection system is capable of spotting and recognizing relationships and similarities between data and also has the capacity to detect anomalies or classify and predict specific events. Deep learn leverages both supervised learning techniques, such as the classification of suspicious transactions, and unsupervised learning, e.g. anomaly detection. The problem confronting the study is to appraise application of deep learning for fraud detection in e-payment system.
1.4 Aim and Objectives of Study
The aim of the study is to appraise the Application of Deep Learning for Fraud Detection in e-Payment System. In achieving this aim, the following specific objectives were laid out as follows:
- To determine the level of fraud in e-payment system.
- To determine the nature and significance of deep learning.
- To determine the effect of the application of deep learning on fraud detection in e-payment system.
1.5 Research Questions
The study came up with research questions so as to be able to ascertain the above stated objectives. The specific research questions for the study are stated below as follows:
- What is the level of fraud in e-payment system?
- What is the nature and significance of deep learning?
- What is the effect of the application of deep learning on fraud detection in e-payment system?
1.6 Research Hypothesis
In order to pursue the objective of this study, the following generalized statements have been designed to guide and aids in obtaining the result for the experiment to be conducted. For this work, the null hypothesis will be represented with H0 while the alternative hypothesis will be represented with hypothesis H1.
Hypothesis One
- H0: The level of fraud in the e-payment system is low.
- H1: The level of fraud in the e-payment system is high.
Hypothesis Two
- H0: The effect of the application of deep learning on fraud detection in e-payment system is negative.
- H1: The effect of the application of deep learning on fraud detection in e-payment system is positive.
1.7 Significance of Study
The study calls on relevant stakeholders on the need to adopt stronger measure for the detecting of fraud in e-payment transactions. Consequently, the study proffers an appraisal of deep learning as an appropriate measure for the detection of fraud in e-payment system.
This study will also be of immense benefit to other researchers who intend to know more on this study and can also be used by non-researchers to build more on their research work. This study contributes to knowledge and could serve as a guide for other study.
1.8 Scope of Study
The scope of the research is focused on the Application of Deep Learning for Fraud Detection in e-Payment System.
1.9 Limitations of the Study
During the course of this study, many things militated against its completion, some of which are:
- Time Constraint: The time frame given to accomplish this project was very short due to school academic calendar and it was carried out under pressure which made the researcher not to implement some necessary features.
- Research material: availability of research material is a major setback to the scope of the study.
- Frequent power failure: This made the researcher append more money on fuel to ensure sustainable power.
- Financial Constraint: Insufficient fund tends to impede the efficiency of the researcher in sourcing for the relevant materials, literature or information and in the process of data collection (internet, questionnaire and interview).
1.10 Definition of Terms
Deep Learning Defined: Deep learning is a class of machine learning algorithms that use a cascade of multiple layers of nonlinear processing units for feature extraction and transformation (Deng et al., 2014).
Fraud: Is defined as deception deliberately practiced with a view of gaining an unlawful or unfair advantage.
Economic Crime: This is the manifestation of criminal act done either solely or in an organized manner with or without associates or group with the propose earning wealth or being rich through all illegal means.
Identity Theft: This is an action that proceeds enable a fraud to occur.
False Billing Fraud: This occurs when a business or an individual receives a bill for a product whereby the representation of the product by the promoter was either false or misleading or whereby the product were either never ordered or received.