Project – The Application of Artificial Intelligence in Detecting Online Financial Fraud: A Case Study of Selected Fintech Platforms in Nigeria.

CHAPTER ONE

INTRODUCTION

Research Topic: The Application of Artificial Intelligence in Detecting Online Financial Fraud: A Case Study of Selected Fintech Platforms in Nigeria.

1.1 Background to the Study

The rapid advancement of information and communication technology has significantly transformed the global financial sector, particularly through the development of digital payment systems, electronic banking, mobile money services, and financial technology platforms. Financial transactions that were traditionally conducted through physical banking halls are now increasingly performed through mobile applications, internet banking, digital wallets, payment gateways, and other electronic platforms. These innovations have improved the speed, convenience, accessibility, and efficiency of financial services, allowing individuals and businesses to transfer funds, make payments, access credit facilities, and conduct financial transactions without necessarily visiting conventional banking institutions. However, the expansion of digital financial services has also created opportunities for sophisticated forms of online financial fraud, including identity theft, account takeover, phishing, unauthorised transactions, payment manipulation, and fraudulent transfers. The increasing complexity of these activities has made fraud detection an important responsibility for financial institutions and technology providers. Bolton and Hand (2002) explain that financial fraud detection involves identifying unusual patterns and suspicious activities within financial transactions, while Abdallah, Maarof, and Zainal (2016) emphasise the importance of developing effective detection techniques to address the changing nature of fraudulent activities. Consequently, the need for innovative technologies capable of identifying and preventing online financial fraud has become increasingly important in the contemporary financial system.

Artificial Intelligence (AI) has emerged as an important technological innovation with considerable potential for improving fraud detection and financial security. Artificial intelligence refers to the development of computer systems capable of performing tasks that ordinarily require human intelligence, including learning, reasoning, pattern recognition, prediction, and decision-making. In financial services, AI can analyse large volumes of transactional information, identify unusual behavioural patterns, detect suspicious activities, and generate alerts that assist financial institutions in preventing fraudulent transactions. Unlike conventional fraud detection approaches that often depend on predetermined rules and manual investigations, AI-based systems can learn from historical transaction records and identify patterns associated with previously observed fraudulent activities. Machine learning, deep learning, natural language processing, and predictive analytics are among the technologies used in developing intelligent fraud detection systems. Ngai, Hu, Wong, Chen, and Sun (2011), in their review of data mining techniques for financial fraud detection, establish that intelligent analytical methods provide important opportunities for identifying fraudulent patterns across financial activities. Similarly, West and Bhattacharya (2016) explain that intelligent financial fraud detection systems can improve the identification of suspicious transactions by applying computational techniques to complex financial data. Therefore, artificial intelligence offers an important technological approach to addressing the growing challenges associated with online financial fraud.

The application of artificial intelligence in financial fraud detection is particularly important because fraudulent transactions frequently involve patterns that may be difficult to identify through conventional monitoring systems. Traditional fraud detection mechanisms commonly rely on fixed transaction limits, predefined rules, manual verification, and alerts generated when transactions violate established security conditions. Although these approaches remain useful, fraudsters continually develop new techniques to bypass security controls, manipulate transaction processes, and exploit weaknesses in digital payment systems. AI-powered fraud detection systems can address some of these limitations by analysing transaction frequency, transaction amounts, geographical locations, device information, account behaviour, and other relevant indicators to identify suspicious activities. Machine learning algorithms can also be trained using historical datasets containing legitimate and fraudulent transactions to distinguish between normal and abnormal financial behaviour. Jurgovsky et al. (2018), in their study of sequence classification for credit-card fraud detection, demonstrate the relevance of sequential learning approaches in identifying fraudulent transaction patterns. Furthermore, Bahnsen, Aouada, and Ottersten (2016) explain that effective feature engineering improves the performance of machine learning models in credit-card fraud detection. These developments demonstrate that AI can support more responsive and data-driven approaches to identifying suspicious financial transactions.

Another important application of artificial intelligence in fraud detection is its ability to support real-time transaction monitoring and automated risk assessment. In digital financial environments, transactions may occur within seconds, making it necessary for fintech platforms to identify suspicious activities before funds are transferred or withdrawn. AI systems can analyse transactions as they occur and assign risk scores based on patterns associated with fraudulent behaviour. Transactions considered suspicious may be flagged for further investigation, subjected to additional verification, delayed, or temporarily restricted in accordance with the platform’s security procedures. Deep learning and anomaly detection techniques can also identify unusual transaction patterns that differ significantly from a customer’s normal financial behaviour. West and Bhattacharya (2016) identify the value of intelligent systems in detecting financial fraud, while Jurgovsky et al. (2018) demonstrate how sequential machine learning methods can improve fraud detection by considering patterns across transaction sequences. More recently, Jejeniwa, Jejeniwa, and Owolabi (2025) investigated a hybrid artificial intelligence approach combining Graph Neural Networks and Isolation Forest anomaly detection using an anonymised dataset from a Nigerian fintech company. Their reported model achieved an F1-score of 0.92 and an area under the receiver operating characteristic curve of 0.98. This illustrates the potential of combining different AI techniques to identify complex and emerging fraudulent transaction patterns.

In Nigeria, the growth of financial technology platforms has significantly changed the delivery and consumption of financial services. Fintech companies provide digital payment solutions, electronic wallets, payment processing, money transfers, lending services, merchant payments, and other technology-based financial products. These services have expanded access to financial transactions and provided individuals and businesses with alternatives to conventional banking channels. However, the increasing volume of digital transactions has also created challenges relating to cybersecurity, fraudulent transfers, account compromise, identity manipulation, and unauthorised access to financial accounts. The Central Bank of Nigeria (CBN, 2024), in its December edition of CBN Update, highlighted the relationship between the expansion of digital payments and increasing cybersecurity concerns within Nigeria’s financial system. The publication identified challenges such as limited cybersecurity awareness, institutional weaknesses, insufficient know-your-customer practices, and inadequate fraud management systems. It also highlighted the potential of artificial intelligence and machine learning in improving fraud detection and strengthening financial system security. Similarly, Edosa, Orukpe, Igori, and Ekoko (2025), in their study of artificial intelligence intervention for predicting fraud and forgeries in the Nigerian banking industry, examined machine learning models using quarterly fraud and forgery records covering 2021 to 2024. Their study demonstrates the growing relevance of artificial intelligence in analysing fraud-related financial data within the Nigerian financial sector.

Fintech platforms operating in Nigeria face different forms of online financial fraud because of the nature, speed, and volume of transactions conducted through their digital systems. Examples of such platforms include mobile payment applications, electronic wallets, payment gateways, digital banking services, and online financial transfer systems. Fraudulent activities may involve phishing, social engineering, fraudulent account registration, identity theft, account takeover, unauthorised withdrawals, suspicious transfers, and the manipulation of payment credentials. These activities can result in financial losses, customer dissatisfaction, reputational damage, and reduced confidence in digital financial services. AI-based fraud detection systems can assist fintech platforms by monitoring transaction behaviour, identifying suspicious account activities, detecting unusual relationships between accounts and devices, and supporting automated risk assessment. However, the effectiveness of these systems may depend on the quality and availability of transaction data, technological infrastructure, algorithmic accuracy, skilled personnel, and the ability to respond to changing fraud patterns. The study by Jejeniwa et al. (2025) demonstrates the potential of hybrid AI models in identifying complex fraud patterns within the Nigerian fintech environment. Similarly, Osa et al. (2025) examined machine learning models for predicting fraud cases and financial losses in the Nigerian banking industry, highlighting the application of predictive analytics in financial fraud management.

Furthermore, the application of artificial intelligence in detecting online financial fraud raises important issues relating to data protection, privacy, algorithmic bias, transparency, accountability, and regulatory compliance. Although AI systems can improve the speed and accuracy of fraud detection, their effectiveness depends on the quality of the data used for training and the appropriateness of the algorithms applied. Poor-quality data may produce inaccurate predictions, while excessive false-positive alerts may cause legitimate transactions to be delayed or rejected, thereby inconveniencing customers. There is also the possibility that sophisticated fraudsters may deliberately manipulate their activities to avoid detection by automated systems. Consequently, fintech platforms must ensure that AI-based fraud detection systems are regularly evaluated, monitored, and improved. The National Institute of Standards and Technology (NIST, 2023), through its Artificial Intelligence Risk Management Framework, emphasises the importance of managing risks associated with the development and deployment of AI systems, particularly concerning reliability, transparency, security, and accountability. In Nigeria, the Nigeria Data Protection Act (2023) establishes requirements for protecting personal data, while the Cybercrimes (Prohibition, Prevention, etc.) Act, 2015, as amended in 2024, provides a legal framework for addressing cybercrime. These considerations demonstrate that successful AI-driven fraud detection requires not only technological innovation but also responsible data management, effective regulation, and appropriate institutional oversight.

Against this background, the present study seeks to examine the application of artificial intelligence in detecting online financial fraud, using selected fintech platforms in Nigeria as its case study. The study recognises the growing importance of digital financial services and the need for reliable technologies capable of identifying suspicious transactions, preventing unauthorised financial activities, and reducing financial losses. It focuses on the AI techniques employed in fraud detection, their effectiveness in identifying suspicious online transactions, their contribution to preventing financial fraud, and the challenges associated with their implementation. It also considers factors such as data quality, infrastructure, implementation costs, technical expertise, algorithmic limitations, and regulatory requirements that may influence the effectiveness of AI-based fraud detection systems. Drawing on the findings of West and Bhattacharya (2016), NIST (2023), Jejeniwa et al. (2025), and Osa et al. (2025), the study seeks to establish how artificial intelligence can strengthen fraud detection mechanisms and improve the security and reliability of selected fintech platforms operating in Nigeria.

1.2 Statement of the Problem

The rapid expansion of digital financial services in Nigeria has created new opportunities for financial inclusion, electronic payments, and convenient financial transactions. However, this development has also increased the exposure of fintech platforms to different forms of online financial fraud, including phishing, identity theft, account takeover, unauthorised transfers, fraudulent account creation, and payment manipulation. These fraudulent activities threaten the security of customers’ funds, undermine confidence in digital financial services, and expose fintech companies to financial and reputational losses. Traditional fraud detection methods, which often depend on predefined transaction rules, manual investigations, and conventional verification procedures, may experience difficulties identifying sophisticated and rapidly changing fraud patterns. Bolton and Hand (2002) explain that financial fraud detection requires the identification of unusual patterns within large volumes of transaction data, while Abdallah et al. (2016) highlight the need for improved detection techniques to address evolving fraudulent activities. Consequently, there is a growing need for more advanced and adaptable technologies capable of identifying suspicious financial transactions within digital financial environments.

Another major problem concerns the limitations of conventional fraud detection mechanisms in responding to the increasing complexity and speed of online financial transactions. As fintech platforms process large volumes of transactions within short periods, manual monitoring and rigid rule-based systems may struggle to identify suspicious activities before financial losses occur. Fraudsters may also modify their transaction patterns, use multiple accounts, manipulate identity information, and exploit weaknesses in digital payment systems to avoid detection. Although artificial intelligence offers opportunities for automated transaction monitoring, predictive analysis, anomaly detection, and behavioural assessment, its effectiveness depends on the availability of reliable data, appropriate algorithms, and adequate technological infrastructure. Jejeniwa et al. (2025) demonstrate the potential of hybrid AI models for identifying fraudulent transactions within the Nigerian fintech environment. However, their study also illustrates the importance of addressing challenges associated with data availability, evolving fraud patterns, and the practical integration of intelligent detection systems. Therefore, the extent to which selected fintech platforms can effectively utilise AI technologies to address online financial fraud requires further investigation.

Furthermore, the application of artificial intelligence in fraud detection may be affected by operational, financial, technical, and regulatory challenges. Some fintech platforms may face high implementation and maintenance costs, inadequate technical expertise, poor-quality transactional data, and difficulties integrating AI systems with existing financial infrastructure. AI-based systems may also generate false-positive results by incorrectly classifying legitimate transactions as fraudulent, thereby causing unnecessary transaction restrictions and customer dissatisfaction. In addition, concerns about customer privacy, data protection, algorithmic transparency, and accountability may affect the adoption and management of AI technologies. The National Institute of Standards and Technology (2023) identifies the importance of managing AI-related risks to ensure that artificial intelligence systems remain reliable, secure, transparent, and accountable. Similarly, the Central Bank of Nigeria (2024) identifies inadequate fraud management tools, institutional weaknesses, and limited technical capacity as concerns within Nigeria’s digital financial environment. These challenges raise questions about the readiness of selected fintech platforms to implement and sustain effective AI-powered fraud detection systems.

The specific problem addressed by this study is the need for stronger empirical understanding of how artificial intelligence is being applied to detect online financial fraud and the extent to which these technologies improve fraud detection effectiveness among selected fintech platforms in Nigeria. Although previous studies, such as West and Bhattacharya (2016), have examined intelligent financial fraud detection techniques, and Jejeniwa et al. (2025) have investigated hybrid AI approaches within the Nigerian fintech environment, there remains a need to examine the application, effectiveness, and implementation challenges of AI-based fraud detection across selected fintech platforms. Edosa et al. (2025) also examined machine learning applications in predicting fraud and financial losses within the Nigerian banking industry, but their focus was on the prediction of fraud-related financial outcomes rather than the broader operational application of AI in fintech fraud detection. Consequently, this study seeks to examine the AI technologies used by selected fintech platforms, assess their effectiveness in identifying suspicious transactions, determine their contribution to preventing online financial fraud, and identify the challenges affecting their implementation in Nigeria.

1.3 Objectives of the Study

The main objective of this study is to examine the application of artificial intelligence in detecting online financial fraud, using selected fintech platforms in Nigeria as a case study.

The specific objectives are to:

1. Examine the artificial intelligence technologies and techniques employed by selected fintech platforms in Nigeria for detecting online financial fraud.

2. Assess the effectiveness of artificial intelligence in identifying suspicious transactions and detecting fraudulent financial activities.

3. Determine the contribution of artificial intelligence to the prevention and reduction of online financial fraud among selected fintech platforms in Nigeria.

4. Identify the challenges affecting the implementation and effective utilisation of artificial intelligence in detecting online financial fraud among selected fintech platforms in Nigeria.

1.4 Research Questions

The following research questions are formulated to guide the study:

1. What artificial intelligence technologies and techniques are employed by selected fintech platforms in Nigeria for detecting online financial fraud?

2. How effective is artificial intelligence in identifying suspicious transactions and detecting fraudulent financial activities?

3. To what extent does artificial intelligence contribute to the prevention and reduction of online financial fraud among selected fintech platforms in Nigeria?

4. What challenges affect the implementation and effective utilisation of artificial intelligence in detecting online financial fraud among selected fintech platforms in Nigeria?

1.5 Research Hypothesis

The following null hypothesis is formulated to guide the study:

H₀₁: There is no significant relationship between the application of artificial intelligence and the effectiveness of online financial fraud detection among selected fintech platforms in Nigeria.

1.6 Significance of the Study

This study is expected to be significant to fintech companies, financial institutions, customers, government regulatory agencies, cybersecurity professionals, policymakers, and future researchers. Its findings will provide useful information on the importance of artificial intelligence in strengthening financial security and addressing the challenges associated with online financial fraud.

1. Fintech Companies: The study will benefit fintech companies by providing information on how artificial intelligence can improve their fraud detection systems. It will help them understand the importance of adopting machine learning, anomaly detection, predictive analytics, and automated transaction monitoring to identify suspicious financial activities. The findings may also assist fintech operators in improving their technological infrastructure and fraud prevention strategies.

2. Financial Institutions: The study will provide useful information to banks, payment service providers, and other financial institutions concerning the application of artificial intelligence in detecting fraudulent transactions. It may encourage financial institutions to strengthen their collaboration with fintech companies in developing more effective fraud monitoring and prevention systems.

3. Customers and Users of Fintech Platforms: The study will benefit customers by highlighting the importance of technological innovations in protecting their financial transactions and digital accounts. Improved fraud detection systems may help reduce unauthorised transactions, financial losses, identity theft, and other forms of online financial exploitation.

4. Government and Regulatory Agencies: The findings will be useful to the Central Bank of Nigeria (CBN), the Nigeria Data Protection Commission (NDPC), and other relevant regulatory institutions. The study may provide information that can support the development of appropriate policies and regulatory guidelines for the responsible application of artificial intelligence in Nigeria’s financial technology sector.

5. Cybersecurity Professionals: The study will provide cybersecurity experts and information technology professionals with useful insights into the role of artificial intelligence in identifying emerging financial fraud patterns. It may encourage the development of more advanced security systems capable of responding to changing cybercrime techniques.

6. Policymakers: The study will provide information that may guide policies concerning digital financial security, artificial intelligence adoption, financial technology innovation, consumer protection, and data governance in Nigeria.

7. Future Researchers: The study will serve as reference material for students and researchers interested in artificial intelligence, financial technology, cybersecurity, online financial fraud, and digital payment security. It will also provide a foundation for further studies on the application of emerging technologies in financial fraud prevention.

1.7 Scope of the Study

This study focuses on the application of artificial intelligence in detecting online financial fraud, using selected fintech platforms in Nigeria as a case study.

Geographically, the study covers selected fintech platforms operating within Nigeria. The investigation is concerned with their application of artificial intelligence in monitoring digital transactions and identifying fraudulent financial activities.

In terms of content, the study examines four major areas:

1. Artificial intelligence technologies employed in detecting online financial fraud, including machine learning, deep learning, anomaly detection, predictive analytics, and automated transaction monitoring.

2. The effectiveness of artificial intelligence in identifying suspicious transactions, detecting unusual financial patterns, and recognising fraudulent activities.

3. The contribution of artificial intelligence to preventing and reducing online financial fraud, including the identification of suspicious accounts, transaction monitoring, and automated fraud alerts.

4. The challenges affecting the implementation of artificial intelligence, including inadequate funding, poor-quality data, technological limitations, shortage of skilled personnel, algorithmic bias, and regulatory concerns.

The study will focus on selected fintech platforms and relevant personnel involved in information technology, cybersecurity, risk management, fraud monitoring, and financial operations, subject to access and the final research design.

1.8 Operational Definition of Terms

The following terms are defined according to their usage in this study.

Artificial Intelligence (AI): This refers to computer-based technologies designed to perform tasks that ordinarily require human intelligence, including learning, reasoning, prediction, pattern recognition, and decision-making.

Financial Fraud: This refers to dishonest or illegal activities involving financial transactions, accounts, payment systems, or financial information for the purpose of obtaining money or other financial benefits unlawfully.

Online Financial Fraud: This refers to fraudulent activities carried out through internet-based financial platforms, electronic payment systems, mobile applications, digital wallets, and other online financial services.

Fraud Detection: This refers to the process of identifying suspicious transactions, unusual financial activities, and patterns that may indicate fraudulent behaviour.

Fintech: This refers to the application of digital technologies and innovative software solutions in providing financial services such as electronic payments, money transfers, digital banking, lending, and financial transactions.

Machine Learning: This refers to a branch of artificial intelligence that enables computer systems to learn from data, recognise patterns, and make predictions or decisions without being explicitly programmed for every situation.

Deep Learning: This refers to an advanced form of machine learning that uses artificial neural networks with multiple layers to analyse complex data and identify patterns.

Anomaly Detection: This refers to the process of identifying unusual or abnormal transaction patterns that differ significantly from expected financial behaviour.

Predictive Analytics: This refers to the use of historical data, statistical techniques, and artificial intelligence algorithms to predict possible future events, including suspicious financial activities.

Transaction Monitoring: This refers to the continuous observation and analysis of financial transactions to identify unusual activities or potential fraudulent behaviour.

Fraud Prevention: This refers to the strategies, technologies, and procedures employed to stop or reduce fraudulent activities before financial losses occur.

Digital Payment Platforms: These refer to electronic systems and applications that enable individuals and organisations to make, receive, and process financial payments through digital channels.

False Positive: This refers to a situation where a legitimate financial transaction is incorrectly identified by a fraud detection system as fraudulent or suspicious.

Project – The Application of Artificial Intelligence in Detecting Online Financial Fraud: A Case Study of Selected Fintech Platforms in Nigeria.