Fraud Detection in Fintech Leveraging Machine Lear
Fraud Detection in Fintech Leveraging Machine Lear
Research Article
Keywords:
DOI: https://doi.org/10.21203/rs.3.rs-3548343/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License
Page 1/23
Abstract
Fraud detection in the fintech sector is a critical area of concern as financial transactions increasingly
shift to digital platforms. This paper presents a comprehensive analysis of enhancing fraud detection in
fintech by combining machine learning techniques, leveraging behavioral analytics, and adopting
RegTech solutions. The objective is to develop a holistic approach that strengthens fraud prevention
strategies, ensures regulatory compliance, and safeguards the interests of customers and financial
institutions. The paper begins with an introduction that sets the context by highlighting the growing
importance of fraud detection in the digital financial landscape. It outlines the research objectives, scope,
and structure of the paper. Subsequently, the methodology section details the data collection process, the
selection and comparative analysis of machine learning models, the integration of behavioral analytics,
and the implementation of RegTech solutions. The paper concludes with a summary of findings and
contributions, emphasizing the significance of adopting a holistic approach to fraud detection in the
fintech industry. It underscores the need for financial institutions to embrace advanced technologies,
comply with data privacy regulations, and collaborate within the industry to combat financial crimes
effectively.
1.0 Introduction
The fintech industry has witnessed remarkable growth and disruption in recent years, reshaping the
landscape of financial services worldwide. With innovative platforms offering seamless digital
experiences, fintech has revolutionized traditional banking, payments, investment, and lending processes
(Lee & Shin, 2018). Giglio (2021) noted that this rapid advancement has also brought forth new
challenges, especially in the realm of fraud and security.
According to Davradakis and Santos (2019), financial fraud has always been a significant concern for
both traditional financial institutions and fintech companies. The rise of digital transactions and the
increasing reliance on technology have provided fraudsters with new avenues to exploit vulnerabilities in
the system. As fintech platforms handle large volumes of sensitive financial data and facilitate numerous
transactions daily, they become prime targets for fraudulent activities (Beck, 2020). Moreover, the fast-
paced nature of fintech operations demands real-time fraud detection and prevention capabilities,
necessitating advanced tools and methodologies.
Conversely, traditional fraud detection methods, based on static rule-based systems, have proven to be
insufficient in combating the ever-evolving nature of fraud (Nicholls et al., 2021). These rule-based
approaches are limited in their ability to adapt to dynamic fraud patterns, leading to higher false positives
and missed detections. As fraudulent schemes become more sophisticated and exploit the vulnerabilities
of traditional methods, there arises a pressing need for more intelligent and adaptive solutions.
1.1 Background
Page 2/23
Amid the growing threats of fraud in the fintech sector, researchers and industry practitioners have been
exploring innovative approaches to enhance fraud detection capabilities (Sanni, 2019; Sengupta et al.,
2020). Among the prominent methodologies being investigated are Machine Learning Techniques and
Algorithms, Behavioral Analytics, and RegTech Solutions.
Machine learning techniques, driven by advancements in artificial intelligence, have shown significant
promise in detecting fraudulent activities in various domains (Blasch et al., 2021). By training on large
datasets of historical transactional information, machine learning models can learn patterns and
anomalies associated with fraud. Mishra and Tyagi (2022) claim that these models can then classify
incoming transactions in real time, identifying suspicious activities and minimizing false positives. The
flexibility and adaptability of machine learning algorithms make them valuable tools in the fight against
fraud in the fintech sector.
On the other hand, Behavioral analytics is another emerging area that offers valuable insights into fraud
detection. Fintech platforms, often engaging with a diverse user base, generate extensive Behavioral data
that can be leveraged to detect anomalies and deviations from regular user Behavior (Pourhabibi et al.,
2020). By analyzing user interactions, navigation patterns, and transactional Behaviors, Behavioral
analytics can pinpoint potential fraud indicators that traditional methods may overlook. Integrating
Behavioral data with fraud detection models can improve accuracy and provide a more comprehensive
understanding of user activities (Deng & Hooi, 2021).
Additionally, RegTech (Regulatory Technology) solutions have emerged as vital components in managing
fraud risks within the fintech industry (Anagnostopoulos, 2018). Regulatory compliance is crucial to
ensure the integrity and stability of financial systems. Fintech companies must adhere to complex and
evolving regulations to prevent financial crimes and protect consumer interests. RegTech solutions offer
automated and streamlined approaches to managing compliance requirements, enabling financial
institutions to focus on identifying and preventing fraudulent activities (Von Solms, 2021).
While each of these methodologies has demonstrated its efficacy in isolation, researchers have
recognized the potential benefits of combining them to create a more robust and comprehensive fraud
detection system. The integration of Machine Learning Techniques, Behavioral Analytics, and RegTech
Solutions can lead to enhanced accuracy, real-time monitoring, and improved risk management.
2.0 Methodology
To achieve the research objectives and explore the integration of Machine Learning Techniques,
Behavioral Analytics, and RegTech Solutions for fraud detection in the fintech sector, a structured and
comprehensive methodology was adopted. The methodology encompasses four key aspects: Data
Collection, Machine Learning Model Selection and Comparative Analysis, Integration of Behavioral
Analytics, and Implementation of RegTech Solutions.
Real-world applications of machine learning in fraud detection span across various fintech domains.
Credit card fraud detection is one of the most prevalent use cases, where machine learning algorithms
analyze transactional patterns and user behavior to identify suspicious activities in real time (Dornadula
& Geetha, 2019). Mytnyk et al. (2023) added that machine learning is also widely employed in preventing
account takeovers, identifying fraudulent account openings, and detecting payment fraud in online
platforms. Meanwhile, Peer-to-peer (P2P) lending platforms use these algorithms to assess borrower
behavior and transactional patterns, reducing the risk of fraudulent loan applications (Jiang et al., 2018).
Additionally, in cryptocurrency and blockchain-based fintech platforms, machine learning techniques
analyze transactional data to detect crypto scams and fraudulent schemes, safeguarding user assets in
the decentralized financial ecosystem (Pocher et al., 2023).
At the heart of behavioral analytics is the understanding of user behavior patterns. Users exhibit
consistent behavior patterns during their interactions with fintech platforms, and these patterns can
provide valuable clues about their typical activities (Claessens et al., 2018). For example, users may have
regular login times, preferred transaction amounts, and specific transaction frequencies. Behavioral
analytics seeks to establish a baseline of normal behavior for each user by analyzing historical data. Any
deviations from this baseline can be indicative of suspicious activities that warrant further investigation.
Page 5/23
A subset of behavioral analytics is behavioral biometrics, which focuses on using unique behavioral
patterns as a means of user identification and authentication (Pfeuffer et al., 2019). Behavioral
biometrics measures biologically inherent traits such as keystroke dynamics, mouse movement patterns,
and touch gestures on mobile devices. These patterns are individualistic and difficult for fraudsters to
mimic, providing an additional layer of security for user authentication (Ikuesan & Venter, 2019). By
leveraging behavioral biometrics, financial institutions can enhance user authentication processes and
prevent unauthorized access to user accounts.
Papantoniou (2022) emphasized that understanding the intricate and ever-evolving regulatory landscape
is paramount for financial institutions. Compliance with anti-money laundering (AML) regulations, know-
your-customer (KYC) requirements, and other regulatory guidelines is essential to maintain the integrity of
financial systems and prevent fraudulent activities. RegTech solutions are designed to keep financial
institutions up-to-date with changing regulations and facilitate compliance automation (Singh et al.,
2021). By automating compliance processes, RegTech solutions reduce the risk of human errors and
ensure that financial institutions stay in line with regulatory obligations.
Moreover, RegTech solutions streamline the KYC verification process, which is essential for ensuring the
identity of customers and meeting regulatory compliance standards (Arner et al., 2019). Manual KYC
verification can be time-consuming and prone to errors, leading to delays in onboarding new customers
and impacting the overall customer experience. RegTech solutions offer automated KYC verification,
reducing manual efforts and expediting the onboarding process for new customers (Eduardo Demarco,
2020). This implies that the automation of KYC verification in financial institutions can ensure a
seamless and efficient onboarding experience for their customers while meeting regulatory requirements.
Page 6/23
data preprocessing and feature engineering, comparative analysis of machine learning models, the use of
ensemble methods for improved accuracy, and the implementation of real-time fraud detection.
Additionally, data cleaning is the initial step in data preprocessing, focusing on identifying and rectifying
errors and inconsistencies in the data (Singh & Dwivedi, 2020). This process involves tasks like removing
duplicate records, correcting erroneous entries, and resolving discrepancies between data sources.
Missing values are a common challenge in real-world datasets, and various techniques can be employed
to handle them (Yadav & Roychoudhury, 2018). Imputation methods, such as mean, median, or mode
imputation, replace missing values with appropriate estimates. Alternatively, techniques like forward fill or
backward fill can be used to propagate the last observed value to fill missing entries in time series data
(Che et al., 2018).
In the context of fraud detection, accurate and reliable model evaluation is of utmost importance. The
consequences of misclassification can be severe, leading to financial losses, reputational damage, and
customer distrust (Chiew et al., 2018). By evaluating the performance of different machine learning
models, financial institutions can identify the ones that achieve the highest level of accuracy in
distinguishing between fraudulent and legitimate transactions. This evaluation process provides valuable
insights into the strengths and weaknesses of each model, enabling data scientists and fraud analysts to
make informed decisions in their fraud risk management strategies (Subrahmanya et al., 2018).
Page 7/23
To measure the performance of machine learning models in fraud detection, various evaluation metrics
are commonly used. These metrics include accuracy, precision, recall (sensitivity), F1-score, and the
Receiver Operating Characteristic (ROC) curve (Cho et al., 2021). Accuracy measures the proportion of
correctly classified instances out of the total instances and provides a fundamental indicator of overall
model performance. However, when dealing with imbalanced data, where the number of legitimate
transactions far outweighs fraudulent ones, accuracy alone may not be the most informative metric
(Mqadi et al., 2021). In such cases, precision, recall, and F1-score become more relevant as they take into
account false positives and false negatives, which are critical in fraud detection.
Boosting, on the other hand, focuses on improving the performance of weak learners by training them
sequentially (Huang et al., 2018). In each iteration, the model gives more weight to misclassified
instances, allowing subsequent models to focus on correcting the mistakes of the previous ones.
AdaBoost and Gradient Boosting Machines (GBM) are well-known boosting algorithms commonly used
in fraud detection (Tama & Rhee, 2019). Boosting helps in identifying difficult-to-detect fraudulent
transactions by iteratively refining the model's ability to distinguish between fraudulent and legitimate
activities.
User behavior patterns are essential in fraud detection for several reasons. They provide a unique and
dynamic view of each individual customer, enabling financial institutions to differentiate between
legitimate users and fraudsters (Feng et al., 2020). By monitoring the regular patterns of a user's login
times, transaction frequency, and typical transaction amounts, the system can identify any deviations
that might indicate suspicious behavior. Understanding these behavioral patterns allows financial
institutions to build a comprehensive profile of each customer, allowing for more accurate and
personalized fraud detection (Chalapathy & Chawla, 2019).
Moreover, user behavior patterns add an additional layer of security to authentication processes.
Traditional methods of authentication, such as passwords or PINs, can be compromised through
phishing attacks or data breaches (Funde et al., 2019). By combining behavioral analytics, including
behavioral biometrics, financial institutions can enhance the security of their authentication systems.
Behavioral biometrics, such as keystroke dynamics or touchscreen interactions, provide unique identifiers
for each user, making it challenging for fraudsters to impersonate legitimate customers (Reddy &
Bodepudi, 2022). The dynamic nature of Behavioral biometrics also ensures that the authentication
process adapts to changes in the user's behavior, further strengthening security.
5.2 Feature Extraction from Behavioral Data
Feature extraction from Behavioral data is a critical step in leveraging user behavior patterns for fraud
detection in the fintech sector. In the context of fraud prevention, Behavioral data refers to the digital
footprints left by customers while using online banking platforms, mobile apps, or other digital financial
services (Pourhabibi et al., 2020). Extracting meaningful features from this data allows financial
institutions to represent user Behavior in a structured format, facilitating the application of machine
learning algorithms for accurate fraud detection.
User Behavior patterns are inherently complex and dynamic, making the extraction of relevant features
crucial for building effective fraud detection models (Wang et al., 2018). The vast amount of Behavioral
data generated by customers while conducting financial transactions necessitates a systematic
approach to extract informative and discriminative features (Marpaung et al., 2021). Features are specific
characteristics or attributes derived from Behavioral data that capture relevant information about user
interactions. These features serve as input variables for machine learning algorithms, enabling them to
identify patterns and make predictions about whether a given activity is legitimate or fraudulent.
Page 9/23
Furthermore, Behavioral biometrics play a significant role in feature extraction. Behavioral biometrics
involves extracting features from unique user Behaviors, such as keystroke dynamics, mouse movement
patterns, touchscreen interactions, and even the way a user holds their smartphone (Reddy & Bodepudi,
2022). By incorporating Behavioral biometrics into the feature extraction process, financial institutions
can enhance the accuracy and reliability of their fraud detection systems. These biometric features add
an extra layer of security to authentication processes, making it harder for fraudsters to impersonate
legitimate customers (Estrela et al., 2021).
Behavioral biometrics analyzes and measures distinct Behavioral patterns exhibited by individuals while
interacting with digital devices (Stylios et al., 2021). These patterns encompass a wide range of
Behavioral characteristics, such as keystroke dynamics, mouse movements, touchscreen interactions,
device handling, and even the manner in which individuals navigate user interfaces. Each person's
Behavior is highly unique, creating a biometric signature that is inherently difficult to imitate or
counterfeit. Unlike traditional biometric modalities like fingerprints or facial recognition, Behavioral
biometrics do not rely on static physical attributes (Liang et al., 2020). Instead, they capture dynamic
Behavioral traits that evolve over time, adapting to changes in the individual's Behavior as they interact
with digital platforms.
Behavioral analytics plays a crucial role in the fraud detection process by establishing a baseline of
normal Behavior for each user (Sharma et al., 2020). This involves analyzing historical data and
Page 10/23
capturing the unique Behavioral patterns exhibited by legitimate users during their interactions with
digital financial services. Integrating Behavioral analytics with machine learning involves the use of both
supervised and unsupervised learning techniques (Zhang et al., 2022). In supervised learning, the model
is trained on labelled data, where historical instances of fraud and non-fraudulent activities are explicitly
identified. The model learns from these labelled examples and uses them to make predictions on new,
unlabeled data. Supervised learning can be particularly useful for building models to detect known types
of fraud and to classify transactions or activities as legitimate or suspicious (Chen et al., 2018).
The primary objective of RegTech is to facilitate compliance with regulatory guidelines, reducing the
manual effort and costs associated with meeting these requirements (Ryan et al., 2020). By automating
data collection, analysis, and reporting, RegTech empowers financial institutions to stay up-to-date with
regulations, identify potential compliance gaps, and demonstrate adherence to regulators effectively. One
of the critical roles of RegTech is to enhance fraud prevention measures within the financial industry
(Buckley et al., 2020). As digital financial services become more prevalent, fraudsters also adapt their
tactics to exploit vulnerabilities in the system. However, RegTech solutions step in to proactively detect
and prevent fraudulent activities through various mechanisms.
Moreover, the manual nature of compliance processes makes them susceptible to errors, delays, and
inconsistencies (Younggren et al., 2022). Human errors in data entry or interpretation can lead to
inaccurate reporting, exposing financial institutions to compliance violations and potential penalties. The
time-consuming nature of manual compliance can divert valuable resources away from strategic
initiatives and proactive risk management efforts. Compliance automation through RegTech solutions
offers financial institutions a transformative approach to addressing the challenges of manual
compliance processes (Arner et al., 2018). RegTech platforms are equipped with advanced technologies,
such as artificial intelligence and data analytics, to automate data collection, analysis, and reporting
tasks.
6.3 Monitoring and Alerts
In the rapidly evolving world of finance, real-time monitoring and alerts have become indispensable tools
for financial institutions to stay ahead of potential risks and respond swiftly to emerging threats (Mughal,
2022). Monitoring and alerts, especially when combined with advanced technologies like artificial
intelligence and machine learning, offer a proactive approach to risk management, fraud prevention, and
compliance adherence. Traditional batch processing or periodic monitoring may introduce delays
between the occurrence of an event and its detection.
In contrast, real-time monitoring ensures that risks are identified as they unfold, reducing the window of
opportunity for fraudsters and mitigating potential losses (Etemadi et al., 2021). Real-time monitoring
and alerts play a critical role in proactive fraud prevention. By integrating Behavioral analytics, machine
learning, and anomaly detection, financial institutions can swiftly identify suspicious activities that may
indicate fraud attempts (Dashottar & Srivastava, 2021). Additionally, real-time monitoring allows financial
institutions to implement dynamic authentication measures. If a transaction or user Behavior appears
suspicious, the system can prompt additional verification steps, such as multi-factor authentication, to
ensure the legitimacy of the activity (Matei-Dimitrie, 2019).
To address data privacy and ethical considerations, financial institutions need to implement robust data
governance and security measures (Janssen et al., 2020). This includes conducting data impact
assessments, obtaining explicit consent for data usage, ensuring data encryption during transmission
and storage, and restricting access to sensitive data to authorized personnel only. By adopting a privacy-
centric approach, financial institutions can build trust with their customers and ensure the responsible
use of data in fraud prevention efforts (Shepherd et al., 2020).
Scalability also ties closely to cost-effectiveness. Implementing and maintaining sophisticated fraud
detection systems can be resource-intensive and expensive. Financial institutions must strike a balance
between investing in advanced technologies and ensuring the cost-effectiveness of their fraud prevention
measures (Akartuna et al., 2022). RegTech solutions can play a vital role in addressing scalability and
cost-effectiveness challenges. By leveraging cloud computing and data virtualization, RegTech platforms
can efficiently scale to handle large workloads while optimizing costs based on actual usage.
Additionally, RegTech solutions offer the advantage of subscription-based models, allowing financial
institutions to pay for the services they use, making it more cost-efficient compared to traditional
infrastructure investments (Preziuso et al., 2022).
Page 13/23
7.3 Adoption and Acceptance in the Fintech Industry
While advanced technologies and RegTech solutions offer significant potential for enhancing fraud
detection and risk management, their successful adoption and acceptance in the fintech industry may
face resistance or challenges (Giudici, 2018). Adoption may be hindered by factors such as
organizational culture, legacy systems, and a lack of awareness or understanding of the benefits of these
technologies. Some financial institutions might be reluctant to adopt new technologies due to concerns
about disrupting existing operations or the perceived complexity of implementation.
To promote adoption and acceptance, financial institutions should invest in comprehensive training and
education programs to familiarize their workforce with the capabilities and advantages of data-driven
fraud prevention solutions (Goh & Wen, 2021). Engaging key stakeholders, including senior management
and compliance teams, in the decision-making process can also foster support for technology
implementation. Collaboration between financial institutions and RegTech providers can also drive
adoption. RegTech vendors can tailor their solutions to meet the specific needs and regulatory
requirements of individual institutions, thus increasing the relevance and attractiveness of the technology
(Maradona & Chand, 2018).
Moreover, the integration of machine learning with Behavioral analytics will create a powerful synergy in
fraud detection. Behavioral biometrics, for instance, will become more refined, enabling financial
institutions to identify users based on unique Behavioral patterns, such as typing speed, mouse
movements, and touchscreen interactions (Gupta et al., 2023). This enhanced level of user authentication
Page 14/23
will strengthen security measures and make it more difficult for fraudsters to deceive traditional
authentication methods.
Edge computing and IoT devices present opportunities for real-time data processing and analysis at the
edge of the network (Sriram, 2022). By deploying fraud detection algorithms on edge devices, financial
institutions can enhance the speed and responsiveness of their fraud prevention efforts, reducing reliance
on centralized servers and minimizing latency. To capitalize on emerging technologies, financial
institutions should conduct pilot projects and proofs of concept to assess their feasibility and
effectiveness (Jain & Mohapatra, 2019). Additionally, staying informed about the latest developments in
emerging technologies and their potential applications in fraud detection will enable financial institutions
to stay ahead of the curve.
9.3 Regulatory Changes and Implications for RegTech
Solutions:
The regulatory landscape governing fraud prevention and risk management is constantly evolving.
Financial institutions must proactively adapt their compliance processes to comply with new regulations
and guidelines (Hassan et al., 2019). RegTech solutions play a pivotal role in helping financial institutions
stay compliant in this ever-changing regulatory environment.
As regulators introduce new requirements, RegTech platforms must be flexible and agile to accommodate
these changes. RegTech providers should maintain close relationships with regulatory authorities to stay
updated on upcoming changes and ensure that their solutions remain compliant (Anagnostopoulos,
2018). Moreover, RegTech solutions should offer robust audit trails and reporting functionalities to
facilitate regulatory examinations. Financial institutions must ensure that their chosen RegTech providers
meet industry standards for data security and privacy to avoid compliance risks.
9. Conclusion
9.1 Summary of Findings:
The paper has explored the various aspects of enhancing fraud detection in the fintech sector through a
comprehensive analysis of machine learning techniques, leveraging Behavioral analytics, and adopting
Page 15/23
RegTech solutions. It highlighted the significance of combining these approaches to create a robust and
proactive fraud prevention strategy for financial institutions. In the section on machine learning
techniques and algorithms, the paper discussed the importance of data preprocessing and feature
engineering in preparing data for analysis. It then delved into the comparative analysis of various
machine learning models, emphasizing the benefits of ensemble methods for improved accuracy and
real-time fraud detection.
The section on leveraging Behavioral analytics for fraud detection explained the value of understanding
user Behavior patterns, extracting meaningful features from Behavioral data, and implementing
Behavioral biometrics for enhanced security. Integrating Behavioral analytics with machine learning
models was emphasized as a powerful means of strengthening fraud detection capabilities.
The discussion on RegTech solutions for fraud risk management highlighted their role in streamlining
compliance processes, enabling real-time monitoring and alerts, and facilitating cross-border transactions
while ensuring regulatory adherence. RegTech was identified as a transformative tool for overcoming
compliance challenges and enhancing risk management practices.
9.2 Significance and Contributions:
This paper makes significant contributions to the understanding of fraud detection in the fintech sector.
By conducting a comparative analysis of machine learning techniques and algorithms, it provides
insights into the most effective approaches for detecting fraudulent activities. It emphasizes the
importance of integrating Behavioral analytics to complement machine learning models, thereby
enhancing fraud prevention capabilities.
The exploration of RegTech solutions' role in fraud risk management demonstrates how technology can
streamline compliance processes and ensure regulatory adherence. The significance of data privacy and
ethical considerations in the use of data-driven technologies for fraud detection is also highlighted,
underscoring the importance of responsible data governance. Overall, this paper contributes to the
growing body of knowledge on fraud detection in fintech and provides valuable recommendations for
financial institutions seeking to strengthen their fraud prevention strategies and compliance measures.
Page 16/23
Furthermore, embracing emerging technologies and staying updated with the latest trends in fraud
detection will enable financial institutions to stay ahead of sophisticated fraudsters. Collaboration and
data-sharing initiatives within the industry will create a united front against financial crimes and enhance
the collective resilience of the fintech sector. However, a holistic approach to fraud detection in the fintech
industry involves combining advanced technologies, responsible data governance, and industry
collaboration. By doing so, financial institutions can bolster their fraud prevention strategies, safeguard
their customers and assets, and foster a safer and more secure financial ecosystem. Embracing
innovation, adopting best practices, and adapting to evolving challenges will be key to staying at the
forefront of fraud detection in the dynamic fintech landscape.
References
1. Akartuna, E. A., Johnson, S. D., & Thornton, A. (2022). Preventing the money laundering and terrorist
financing risks of emerging technologies: An international policy Delphi study. Technological
Forecasting and Social Change, 179, 121632. https://doi.org/10.1016/j.techfore.2022.121632
2. Anagnostopoulos, I. (2018). Fintech and regtech: Impact on regulators and banks. Journal of
Economics and Business, 100, 7-25. https://doi.org/10.1016/j.jeconbus.2018.07.003
3. Arner, D. W., Barberis, J., & Buckley, R. P. (2018). RegTech: Building a better financial system. In
Handbook of blockchain, digital finance, and inclusion, Volume 1 (pp. 359-373). Academic Press.
https://doi.org/10.1016/B978-0-12-810441-5.00016-6
4. Arner, D. W., Zetzsche, D. A., Buckley, R. P., & Barberis, J. N. (2019). The identity challenge in finance:
from analogue identity to digitized identification to digital KYC utilities. European business
organization law review, 20, 55-80. https://doi.org/10.1007/s40804-019-00135-1
5. Ashta, A., & Herrmann, H. (2021). Artificial intelligence and fintech: An overview of opportunities and
risks for banking, investments, and microfinance. Strategic Change, 30(3), 211-222.
https://doi.org/10.1002/jsc.2404
6. Banga, L., & Pillai, S. (2021, July). Impact of Behavioral biometrics on mobile banking system. In
Journal of Physics: Conference Series (Vol. 1964, No. 6, p. 062109). IOP Publishing.
https://doi.org/10.1088/1742-6596/1964/6/062109
7. Beck, T. (2020). Fintech and financial inclusion: Opportunities and pitfalls (No. 1165). ADBI working
paper series. http://hdl.handle.net/10419/238522
8. Blasch, E., Pham, T., Chong, C. Y., Koch, W., Leung, H., Braines, D., & Abdelzaher, T. (2021). Machine
learning/artificial intelligence for sensor data fusion–opportunities and challenges. IEEE Aerospace
and Electronic Systems Magazine, 36(7), 80-93. https://doi.org/10.1109/MAES.2020.3049030
9. Buckley, R. P., Arner, D. W., Zetzsche, D. A., & Weber, R. H. (2020). The road to RegTech: the
(astonishing) example of the European Union. Journal of Banking Regulation, 21, 26-36.
https://doi.org/10.1057/s41261-019-00104-1
10. Chalapathy, R., & Chawla, S. (2019). Deep learning for anomaly detection: A survey. arXiv preprint
arXiv:1901.03407. https://doi.org/10.48550/arXiv.1901.03407
Page 17/23
11. Che, Z., Purushotham, S., Cho, K., Sontag, D., & Liu, Y. (2018). Recurrent neural networks for
multivariate time series with missing values. Scientific reports, 8(1), 6085.
https://doi.org/10.1038/s41598-018-24271-9
12. Chen, Z., Van Khoa, L. D., Teoh, E. N., Nazir, A., Karuppiah, E. K., & Lam, K. S. (2018). Machine learning
techniques for anti-money laundering (AML) solutions in suspicious transaction detection: a review.
Knowledge and Information Systems, 57, 245-285. https://doi.org/10.1007/s10115-017-1144-z
13. Chiew, K. L., Choo, J. S. F., Sze, S. N., & Yong, K. S. (2018). Leverage website favicon to detect
phishing websites. Security and communication networks, 2018.
14. Cho, W. K., Lee, Y. J., Joo, H. A., Jeong, I. S., Choi, Y., Nam, S. Y., ... & Choi, S. H. (2021). Diagnostic
accuracies of laryngeal diseases using a convolutional neural network‐based image classification
system. The Laryngoscope, 131(11), 2558-2566. https://doi.org/10.1002/lary.29595
15. Christin, S., Hervet, É., & Lecomte, N. (2019). Applications for deep learning in ecology. Methods in
Ecology and Evolution, 10(10), 1632-1644. https://doi.org/10.1111/2041-210X.13256
16. Chueca Vergara, C., & Ferruz Agudo, L. (2021). Fintech and sustainability: do they affect each other?.
Sustainability, 13(13), 7012. https://doi.org/10.3390/su13137012
17. Claessens, S., Frost, J., Turner, G., & Zhu, F. (2018). Fintech credit markets around the world: size,
drivers and policy issues. BIS Quarterly Review September. https://ssrn.com/abstract=3288096
18. Dashottar, S., & Srivastava, V. (2021). Corporate banking—risk management, regulatory and reporting
framework in India: A Blockchain application-based approach. Journal of Banking Regulation, 22, 39-
51. https://doi.org/10.1057/s41261-020-00127-z
19. Davradakis, E., & Santos, R. (2019). Blockchain, FinTechs and their relevance for international
financial institutions (No. 2019/01). EIB Working Papers. http://hdl.handle.net/10419/191788
20. Deng, A., & Hooi, B. (2021, May). Graph neural network-based anomaly detection in multivariate time
series. In Proceedings of the AAAI conference on artificial intelligence (Vol. 35, No. 5, pp. 4027-4035).
https://doi.org/10.1609/aaai.v35i5.16523
21. Dornadula, V. N., & Geetha, S. (2019). Credit card fraud detection using machine learning algorithms.
Procedia computer science, 165, 631-641. https://doi.org/10.1016/j.procs.2020.01.057
22. Eduardo Demarco, A. (2020). Analysing blockchain/distributed ledger technology in capital markets
and know your customer process. Journal of Securities Operations & Custody, 12(1), 58-71.
23. Estrela, P. M. A. B., Albuquerque, R. D. O., Amaral, D. M., Giozza, W. F., & Júnior, R. T. D. S. (2021). A
framework for continuous authentication based on touch dynamics biometrics for mobile banking
applications. Sensors, 21(12), 4212. https://doi.org/10.3390/s21124212
24. Etemadi, N., Borbon-Galvez, Y., Strozzi, F., & Etemadi, T. (2021). Supply chain disruption risk
management with blockchain: A dynamic literature review. Information, 12(2), 70.
https://doi.org/10.3390/info12020070
25. Feng, L., Xu, S., Zhang, L., Wu, J., Zhang, J., Chu, C., ... & Shi, H. (2020). Anomaly detection for
electricity consumption in cloud computing: framework, methods, applications, and challenges.
Page 18/23
EURASIP Journal on Wireless Communications and Networking, 2020(1), 1-12.
https://doi.org/10.1186/s13638-020-01807-0
26. Funde, N. A., Dhabu, M. M., Paramasivam, A., & Deshpande, P. S. (2019). Motif-based association rule
mining and clustering technique for determining energy usage patterns for smart meter data.
Sustainable Cities and Society, 46, 101415. https://doi.org/10.1016/j.scs.2018.12.043
27. Giglio, F. (2021). Fintech: A literature review. European Research Studies Journal, 24(2B), 600-627.
28. Giudici, P. (2018). Fintech risk management: A research challenge for artificial intelligence in finance.
Frontiers in Artificial Intelligence, 1, 1. https://doi.org/10.1080/15313220.2020.1768621
29. Goh, E., & Wen, J. (2021). Applying the technology acceptance model to understand hospitality
management students’ intentions to use electronic discussion boards as a learning tool. Journal of
Teaching in Travel & Tourism, 21(2), 142-154. https://doi.org/10.1080/15313220.2020.1768621
30. Gupta, S., Maple, C., Crispo, B., Raja, K., Yautsiukhin, A., & Martinelli, F. (2023). A survey of human-
computer interaction (HCI) & natural habits-based Behavioral biometric modalities for user
recognition schemes. Pattern Recognition, 109453. https://doi.org/10.1016/j.patcog.2023.109453
31. Hassan, M. K., Aliyu, S., Huda, M., & Rashid, M. (2019). A survey on Islamic Finance and accounting
standards. Borsa Istanbul Review, 19, S1-S13. https://doi.org/10.1016/j.bir.2019.07.006
32. Hassan, M. U., Rehmani, M. H., & Chen, J. (2022). Anomaly detection in blockchain networks: A
comprehensive survey. IEEE Communications Surveys & Tutorials.
https://doi.org/10.1109/COMST.2022.3205643
33. Hoofnagle, C. J., Van Der Sloot, B., & Borgesius, F. Z. (2019). The European Union general data
protection regulation: what it is and what it means. Information & Communications Technology Law,
28(1), 65-98. https://doi.org/10.1080/13600834.2019.1573501
34. Hossain, M. J., Rifat, R. H., Mugdho, M. H., Jahan, M., Rasel, A. A., & Rahman, M. A. (2022,
November). Cyber Threats and Scams in FinTech Organizations: A brief overview of financial fraud
cases, future challenges, and recommended solutions in Bangladesh. In 2022 International
Conference on Informatics, Multimedia, Cyber and Information System (ICIMCIS) (pp. 190-195). IEEE.
35. Huang, F., Ash, J., Langford, J., & Schapire, R. (2018, July). Learning deep resnet blocks sequentially
using boosting theory. In International Conference on Machine Learning (pp. 2058-2067). PMLR.
36. Ikuesan, A. R., & Venter, H. S. (2019). Digital behavioral-fingerprint for user attribution in digital
forensics: Are we there yet?. Digital Investigation, 30, 73-89.
https://doi.org/10.1016/j.diin.2019.07.003
37. Jain, K., & Mohapatra, S. (2019). Taxonomy of edge computing: Challenges, opportunities, and data
reduction methods. Edge Computing: From Hype to Reality, 51-69.
38. Janssen, M., Brous, P., Estevez, E., Barbosa, L. S., & Janowski, T. (2020). Data governance: Organizing
data for trustworthy Artificial Intelligence. Government Information Quarterly, 37(3), 101493.
https://doi.org/10.1016/j.giq.2020.101493
39. Jiang, C., Wang, Z., Wang, R., & Ding, Y. (2018). Loan default prediction by combining soft
information extracted from descriptive text in online peer-to-peer lending. Annals of Operations
Page 19/23
Research, 266(1-2), 511-529. https://doi.org/10.1007/s10479-017-2668-z
40. Jurgovsky, J., Granitzer, M., Ziegler, K., Calabretto, S., Portier, P. E., He-Guelton, L., & Caelen, O. (2018).
Sequence classification for credit-card fraud detection. Expert systems with applications, 100, 234-
245. https://doi.org/10.1016/j.eswa.2018.01.037
41. Karplus, V. J., Zhang, J., & Zhao, J. (2021). Navigating and evaluating the labyrinth of environmental
regulation in China. Review of Environmental Economics and Policy, 15(2), 300-322.
42. Khatri, S., Arora, A., & Agrawal, A. P. (2020, January). Supervised machine learning algorithms for
credit card fraud detection: a comparison. In 2020 10th international conference on cloud computing,
data science & engineering (confluence) (pp. 680-683). IEEE.
https://doi.org/10.1109/Confluence47617.2020.9057851
43. Kurum, E. (2023). RegTech solutions and AML compliance: what future for financial crime?. Journal
of Financial Crime, 30(3), 776-794. https://doi.org/10.1108/JFC-04-2020-0051
44. Latif, S., Usman, M., Manzoor, S., Iqbal, W., Qadir, J., Tyson, G., ... & Crowcroft, J. (2020). Leveraging
data science to combat COVID-19: A comprehensive review. IEEE Transactions on Artificial
Intelligence, 1(1), 85-103. https://doi.org/10.1109/TAI.2020.3020521
45. Lee, I., & Shin, Y. J. (2018). Fintech: Ecosystem, business models, investment decisions, and
challenges. Business horizons, 61(1), 35-46. https://doi.org/10.1016/j.bushor.2017.09.003
46. Liang, Y., Samtani, S., Guo, B., & Yu, Z. (2020). Behavioral biometrics for continuous authentication in
the internet-of-things era: An artificial intelligence perspective. IEEE Internet of Things Journal, 7(9),
9128-9143. https://doi.org/10.1016/j.gltp.2022.04.020
47. Maharana, K., Mondal, S., & Nemade, B. (2022). A review: Data pre-processing and data
augmentation techniques. Global Transitions Proceedings, 3(1), 91-99.
https://doi.org/10.1016/j.gltp.2022.04.020
48. Maradona, A. F., & Chand, P. (2018). The pathway of transition to International Financial Reporting
Standards (IFRS) in developing countries: Evidence from Indonesia. Journal of International
Accounting, Auditing and Taxation, 30, 57-68. https://doi.org/10.1016/j.intaccaudtax.2017.12.005
49. Marpaung, F. K., Dewi, R. S., Grace, E., Sudirman, A., & Sugiat, M. (2021). Behavioral Stimulus for
Using Bank Mestika Mobile Banking Services: UTAUT2 Model Perspective. Golden Ratio of Marketing
and Applied Psychology of Business, 1(2), 61-72. https://doi.org/10.52970/grmapb.v1i2.68
50. Matei-Dimitrie, T. (2019). Multi-factor authentication. An extended overview. Romanian Cyber Security
Journal.
51. Mirza, N., Elhoseny, M., Umar, M., & Metawa, N. (2023). Safeguarding FinTech innovations with
Machine Learning: Comparative Assessment of Various Approaches. Research in International
Business and Finance, 102009. https://doi.org/10.1016/j.ribaf.2023.102009
52. Mishra, S., & Tyagi, A. K. (2022). The role of machine learning techniques in internet of things-based
cloud applications. Artificial intelligence-based internet of things systems, 105-135.
53. Mqadi, N. M., Naicker, N., & Adeliyi, T. (2021). Solving misclassification of the credit card imbalance
problem using near miss. Mathematical Problems in Engineering, 2021, 1-16.
Page 20/23
https://doi.org/10.1155/2021/7194728
54. Mughal, A. A. (2022). Building and Securing the Modern Security Operations Center (SOC).
International Journal of Business Intelligence and Big Data Analytics, 5(1), 1-15.
https://orcid.org/0009-0006-8460-8006
55. Mytnyk, B., Tkachyk, O., Shakhovska, N., Fedushko, S., & Syerov, Y. (2023). Application of Artificial
Intelligence for Fraudulent Banking Operations Recognition. Big Data and Cognitive Computing, 7(2),
93. https://doi.org/10.3390/bdcc7020093
56. Nicholls, J., Kuppa, A., & Le-Khac, N. A. (2021). Financial cybercrime: A comprehensive survey of deep
learning approaches to tackle the evolving financial crime landscape. Ieee Access, 9, 163965-
163986. https://doi.org/10.1109/ACCESS.2021.3134076
57. Nigam, D., Patel, S. N., Raj Vincent, P. M., Srinivasan, K., & Arunmozhi, S. (2022). Biometric
authentication for intelligent and privacy-preserving healthcare systems. Journal of Healthcare
Engineering, 2022. https://doi.org/10.1155/2022/1789996
58. Pan, Y., & Zhang, L. (2021). Roles of artificial intelligence in construction engineering and
management: A critical review and future trends. Automation in Construction, 122, 103517.
https://doi.org/10.1016/j.autcon.2020.103517
59. Papadakis, S., Garefalakis, A., Lemonakis, C., Chimonaki, C., & Zopounidis, C. (Eds.). (2020). Machine
Learning Applications for Accounting Disclosure and Fraud Detection. IGI Global.
60. Papantoniou, A. A. (2022). Regtech: steering the regulatory spaceship in the right direction?. Journal
of Banking and Financial Technology, 6(1), 1-16. https://doi.org/10.1007/s42786-022-00038-9
61. Pfeuffer, K., Geiger, M. J., Prange, S., Mecke, L., Buschek, D., & Alt, F. (2019, May). Behavioral
biometrics in VR: Identifying people from body motion and relations in virtual reality. In Proceedings
of the 2019 CHI Conference on Human Factors in Computing Systems (pp. 1-12).
https://doi.org/10.1145/3290605.3300340
62. Pocher, N., Zichichi, M., Merizzi, F., Shafiq, M. Z., & Ferretti, S. (2023). Detecting anomalous
cryptocurrency transactions: An AML/CFT application of machine learning-based forensics.
Electronic Markets, 33(1), 37. https://doi.org/10.1007/s12525-023-00654-3
63. Pourhabibi, T., Ong, K. L., Kam, B. H., & Boo, Y. L. (2020). Fraud detection: A systematic literature
review of graph-based anomaly detection approaches. Decision Support Systems, 133, 113303.
https://doi.org/10.1016/j.dss.2020.113303
64. Preziuso, M., Koefer, F., & Ehrenhard, M. (2023). Open banking and inclusive finance in the European
Union: perspectives from the Dutch stakeholder ecosystem. Financial Innovation, 9(1), 1-27.
https://doi.org/10.1186/s40854-023-00522-1
65. Reddy, M., & Bodepudi, A. (2022). Analysis of Cloud Based Keystroke Dynamics for Behavioral
Biometrics Using Multiclass Machine Learning. ResearchBerg Review of Science and Technology,
2(1), 120-135.
66. Ryan, P., Crane, M., & Brennan, R. (2020). Design challenges for GDPR RegTech. arXiv preprint
arXiv:2005.12138. https://doi.org/10.5220/0009464507870795
Page 21/23
67. Sanni, L. (2019). An Informational Perspective and a Framework for Employee Fraud Detection
(Doctoral dissertation, Université Paris 1 Panthéon Sorbonne, France).
68. Sengupta, S., Basak, S., Saikia, P., Paul, S., Tsalavoutis, V., Atiah, F., ... & Peters, A. (2020). A review of
deep learning with special emphasis on architectures, applications and recent trends. Knowledge-
Based Systems, 194, 105596. https://doi.org/10.1016/j.knosys.2020.105596
69. Sermpezis, P., Kotronis, V., Gigis, P., Dimitropoulos, X., Cicalese, D., King, A., & Dainotti, A. (2018).
ARTEMIS: Neutralizing BGP hijacking within a minute. IEEE/ACM Transactions on Networking, 26(6),
2471-2486. https://doi.org/10.1109/TNET.2018.2869798
70. Hari Prasad Josyula, Deepti Vishnubhotla, Paul Oteyo Onyando (2023). Is Artificial Intelligence an
Efficient Technology for Financial Fraud Risk Management? International Journal of Managerial
Studies and Research, 11(6), 11-16 https://doi.org/10.20431/2349-0349.1106002
71. Sharbek, N. (2022). How Traditional Financial Institutions have adapted to Artificial Intelligence,
Machine Learning and FinTech?. In Proceedings of the International Conference on Business
Excellence (Vol. 16, No. 1, pp. 837-848). https://doi.org/10.1109/ICBATS57792.2023.10111143
72. Sharma, B., Pokharel, P., & Joshi, B. (2020, July). User behavior analytics for anomaly detection using
LSTM autoencoder-insider threat detection. In Proceedings of the 11th international conference on
advances in information technology (pp. 1-9). https://doi.org/10.1145/3406601.3406610
73. Shepherd, M., Turner, J. A., Small, B., & Wheeler, D. (2020). Priorities for science to overcome hurdles
thwarting the full promise of the ‘digital agriculture’revolution. Journal of the Science of Food and
Agriculture, 100(14), 5083-5092. https://doi.org/10.1002/jsfa.9346
74. Shihembetsa, E. (2021). Use of artificial intelligence algorithms to enhance fraud detection in the
Banking Industry (Doctoral dissertation, University of Nairobi).
75. Singh, S. K., & Dwivedi, D. R. K. (2020). Data mining: dirty data and data cleaning. Available at SSRN
3610772. http://dx.doi.org/10.2139/ssrn.3610772
76. Singh, S., & Srivastava, R. K. (2020). Understanding the intention to use mobile banking by existing
online banking customers: an empirical study. Journal of Financial Services Marketing, 25(3-4), 86-
96. https://doi.org/10.1057/s41264-020-00074-w
77. Sriram, G. S. (2022). Edge computing vs. Cloud computing: an overview of big data challenges and
opportunities for large enterprises. International Research Journal of Modernization in Engineering
Technology and Science, 4(1), 1331-1337.
78. Stojanović, B., Božić, J., Hofer-Schmitz, K., Nahrgang, K., Weber, A., Badii, A., ... & Runevic, J. (2021).
Follow the trail: Machine learning for fraud detection in Fintech applications. Sensors, 21(5), 1594.
https://doi.org/10.3390/s21051594
79. Stylios, I., Kokolakis, S., Thanou, O., & Chatzis, S. (2021). Behavioral biometrics & continuous user
authentication on mobile devices: A survey. Information Fusion, 66, 76-99.
https://doi.org/10.1016/j.inffus.2020.08.021
80. Subrahmanya, S. V. G., Shetty, D. K., Patil, V., Hameed, B. Z., Paul, R., Smriti, K., ... & Somani, B. K.
(2022). The role of data science in healthcare advancements: applications, benefits, and future
Page 22/23
prospects. Irish Journal of Medical Science (1971-), 191(4), 1473-1483.
https://doi.org/10.1007/s11845-021-02730-z
81. Tama, B. A., & Rhee, K. H. (2019). An in-depth experimental study of anomaly detection using
gradient boosted machine. Neural Computing and Applications, 31, 955-965.
https://doi.org/10.1007/s00521-017-3128-z
82. Toreini, E., Aitken, M., Coopamootoo, K., Elliott, K., Zelaya, C. G., & Van Moorsel, A. (2020, January).
The relationship between trust in AI and trustworthy machine learning technologies. In Proceedings
of the 2020 conference on fairness, accountability, and transparency (pp. 272-283).
https://doi.org/10.1145/3351095.3372834
83. Vesna, B. A. (2021). Challenges of financial risk management: AI applications. Management: Journal
of Sustainable Business and Management Solutions in Emerging Economies, 26(3), 27-34.
84. Von Solms, J. (2021). Integrating Regulatory Technology (RegTech) into the digital transformation of
a bank Treasury. Journal of Banking Regulation, 22, 152-168. https://doi.org/10.1057/s41261-020-
00134-0
85. Walker, G. A. (2021). Regulatory Technology (Regtech)-Construction of a New Regulatory Policy and
Model. Int'l Law., 54, 1-3.
86. Wang, X., Tan, Q., Shi, J., Su, S., & Wang, M. (2018, June). Insider threat detection using
characterizing user behavior. In 2018 IEEE Third International Conference on Data Science in
Cyberspace (DSC) (pp. 476-482). IEEE. . https://doi.org/10.1109/DSC.2018.00077
87. Sachin Parate, Hari Prasad Josyula, Latha Thamma Reddi (2023) Digital Identity Verification:
Transforming KYC Processes in Banking Through Advanced Technology And Enhanced Security
Measures International Research Journal of Modernization in Engineering Technology and Science,
5(9), 128-137 https://www.doi.org/10.56726/IRJMETS44476
88. Wu, Y., Dai, H. N., & Wang, H. (2020). Convergence of blockchain and edge computing for secure and
scalable IIoT critical infrastructures in industry 4.0. IEEE Internet of Things Journal, 8(4), 2300-2317.
https://doi.org/10.1109/JIOT.2020.3025916
89. Yadav, M. L., & Roychoudhury, B. (2018). Handling missing values: A study of popular imputation
packages in R. Knowledge-Based Systems, 160, 104-118.
https://doi.org/10.1016/j.knosys.2018.06.012
90. Younggren, J. N., Gottlieb, M. C., & Baker, E. (2022). Navigating the labyrinth of professional
regulations: Surviving in a flawed regulatory system. Professional Psychology: Research and
Practice, 53(4), 333.
91. Zhang, G., Li, Z., Huang, J., Wu, J., Zhou, C., Yang, J., & Gao, J. (2022). efraudcom: An e-commerce
fraud detection system via competitive graph neural networks. ACM Transactions on Information
Systems (TOIS), 40(3), 1-29. https://doi.org/10.1145/3474379
92. Zhao, Y., Chen, J., Wu, D., Teng, J., & Yu, S. (2019, December). Multi-task network anomaly detection
using federated learning. In Proceedings of the 10th international symposium on information and
communication technology (pp. 273-279). https://doi.org/10.1145/3368926.3369705
Page 23/23