How to Prevent Fraud: Essay

With the emergence of modern technology, the issues of fraud can now be handled and managed reasonably. Fraud has continued to be the worst problem in the digital world, it certainly has affected a great number of individuals, organizations, institutions, and other aspects. So many measures have been used in managing and controlling fraud, but some of the measures have failed us and this results in a negative effect. In this paper, I am going introduce five latest technologies that can be able to tackle the problem of fraud, namely differential privacy, federated analysis, zero-knowledge proofs, homomorphic encryption, and secure multiparty computation. Although these techniques are still in their initial stage, I am going to explore and see how they can be applied in preventing fraud.

Literature Review

Looking at the fact that much research has been conducted to prevent and defend against fraud, this section will extensively review related works. The paper of E. I. Tarmazakov and D. S. Silnov, 'Modern Approaches to Prevent Fraud in Mobile Communications Networks', presents new ways to combat fraud which is based on mobile communication networks using aggregation of anti-fraud systems. The paper first illustrates the report of the Communication Fraud Control Association CFAC, which explains how a lot of telecommunications company loses billions of dollars yearly due to different types of fraud and methods. A method for mapping data in a communication network to detect phantom subscribers and a method of matching data to security policies were shown. To be able to detect fraudulent events, two methods are to be considered, firstly, monitoring the integrity of data flows will make it easy to detect fraud, and secondly, by analysis of subscriber behaviors, any suspicious activities or events should be detected, the fraud management system will consider all this options using aggregation of anti-fraud systems.

A. El Orche, M. Bahaj, and S. A. Alhayat in their 'Ontology Based on Electronic Payment Fraud Prevention' introduced an approach that applies ontology in protecting and preventing fraud. The advantage of this approach is that there are many important ways to improve electronic payment using rules. The paper starts with an explanation of the meaning of ontology, which is the organized and characterized set of ideas specifying an information field. Its objective is to describe the domain knowledge to be interpreted by both humans and devices. Ontology is concerned with how electronic payment fraud can be limited by sharing rules of fraud with applications used in financial companies and others. The features and importance of electronic payment fraud as well as a detailed explanation of fraud management that includes consequences of fraud, prevention of fraud, fraud risk indicator, and fraud prevention tools.

The paper of N. Balasupramanian, B. G. Ephrem, and I. S. Al-Barwani 'User Pattern Based Online Fraud Detection and Prevention Using Big Data Analytics and Self Organizing Maps' proposed a working model using two techniques that are used in fraud detection and prevention using big data analysis and machine learning using self-organizing mappings. A detailed classification of fraud both online/offline fraud was explained. The fraud detection methods are based on monitoring any suspicious pattern behavior, algorithms that are used in detecting frauds like artificial neural networks algorithms, machine learning algorithms, data mining algorithms, etc. Big data analysis is a method of analyzing data from a very big set to bring out the important data from the set and also involves many techniques like cluster analysis, data mining, genetic algorithms, principal components analysis, and self-organizing maps. The model is split into four components: data collection, preprocessing, training of the data, and mapping of data to verify accuracy. When the pattern matches, transactions will be approved or else not approved.

P. Sarda, M. J. M. Chowdhury, A. Colman, M. A. Kabir, and J. Han in 'Blockchain for Fraud Prevention: A Work-History Fraud Prevention System' proposed a new approach using blockchain concept in fraud prevention based on a work-history (job) fraud prevention system. Blockchain was used to store encrypted types of work history and verifications of the information in an organization. A prototype was developed for the verification process implemented using .Net and Ethereum public blockchain. The requirements analysis was done and it shows how important the prototype will enhance previous work- history verification processes. The implementation was based on smart contracts and employer clients. Two algorithms were used on how data is encrypted and shared work history using blockchain. The strengths of this method are that it reduces the cost of hiring third parties in verification processes and guarantees the integrity and confidentiality of data using blockchain. The limitations to the method are that it is only applicable on the Ethereum public blockchain, and also the keys for the encryption need to be secured because if lost there will be a problem.

An interesting paper of A. Tatiana explores the different classifications of fraud in general. The classification and qualification are based on public sector how fraud is managed and observed, which includes waste, abuse, kickback, accounting fraud, procurement fraud, asset misappropriation, and bribery and corruption. A fraud prevention framework was introduced to help government auditors which consists of budgetary operation/transactions surfing, extraction of suspicious operations, brainstorming to identify and detect fraud, and lastly assessing fraud.

Also, in 'Fraud Analysis and Prevention in E-Commerce Transactions', E. Caldeira, G. Brandao, and A. C. M. Pereira present a case study that is based on some techniques that can be combined to prevent fraud in e-commerce and credit card transactions. Four different computational intelligence techniques were used, including Bayesian networks, logistics regression, neural networks, and random forest. The idea of the case study presented was derived from all the four techniques. A dataset known as PaySeguro was used and the same methodology was adopted for all the techniques with different parameters for each technique. A comparative result for all the techniques based on the dataset used was illustrated.

Save your time!
We can take care of your essay

• Proper editing and formatting
• Free revision, title page, and bibliography
• Flexible prices and money-back guarantee

Place Order

The paper of M. A. R. Vol '& ACCOUNTING' reviews five different computer technology applications that can be used in fraud detection and fraud prevention processes, which are spreadsheets, big data, forensic analytics, text analytics, and expert systems.

O. Avdeyuk, D. Kozlov, L. Druzhinina, and I. Tarasova proposed an approach for monitoring the service work of a POS network in real-time that can be used for analyzing, detecting, and preventing fraud in the system of electronic payments in real-time database using RethinkDB.

C. Guo, H. Wang, H. N. Dai, S. Cheng, and T. Wang put forward a fraud risk monitoring system for e-banking transactions using two models together, which are the model of score rules for online real-time transactions and offline historical transactions. A review of the fraud detection system was explained, including fraud risk detection and fraud risk monitoring evaluation. The system was designed in two models. The first is the RAIB model, which is classified into three parts activity of the information, identity shows the account status and behavior of the customer in fraud monitoring, and a parallel random forest algorithm was developed. In the offline transaction KSMG model, which collects the transaction log files from the e-banking system and stores them, a fraud algorithm was also used to show the process.

Modern Technology to Prevent Fraud

The World Economic Forum has pointed out the latest technologies that will be very effectively and efficiently used in combatting fraud by protecting individuals' and organizations' information, as well as preventing loss. These modern technologies include differential privacy, federated analysis, homomorphic encryption, zero-knowledge proofs, and secure multiparty computation.

1. Differential privacy. It is a technique that can be able to prevent fraud by protecting the privacy of confidential information of individuals or organizations from being accessed by fraudsters. It is a mathematical technique that uses noise or sound in the set of information inside the system to alarm any suspicious act by preventing data breaches that can lead to fraud.
2. Federated analysis. It is a technique used to put together the information collectively from other systems into a single system for easy access and analysis. The technique can make fraud monitoring and detection at ease by quickly analyzing information, which will lead to the pronouncement of fraud and increased security because important information can be exposed and detected.
3. Homomorphic encryption. It is a technique that encrypts information that can be analyzed and processed in a such that the information is concealed and protected during the process. This technique can strongly protect against fraud in transactions due to the fact the information will not be decrypted when processing. It is a promising technique that will prevent individuals and organizations from falling victim to fraud by controlling and protecting the information. For example, even if the fraudsters access the information, it cannot be of benefit because it is already encrypted.
4. Zero-knowledge proofs. It is a technique that can be applied to prevent fraud. It is by hiding the actual information that the individuals or organization doesn’t want to reveal, but also at the same time sharing some part of the information that is relevant.
5. Secure multiparty computation. It is a technique that permits individuals or organizations to be involved in an operation collectively, but each other protects its confidential information.

Importance of Modern Technologies in Preventing Fraud

The evolvement of different fraud techniques has brought about great concern in preventing them. The following are the importance of modern technologies in curbing the threat of fraud in an organization. Firstly, improvement in the prevention of fraudulent schemes, for example, the differential privacy technique, has an advantage that even when the fraudsters access the information, they cannot be able to use it because of the noise added to it. Another fact is that federated analysis allows information sharing to only intended parties and also prevents sensitive information from being exposed. This idea will definitely reduce fraud.

Advanced security by verification before sharing information with other parties, for example, zero-knowledge proofs, can check up information before revealing it; this will enhance the security of information and at the same time limit the effect of fraud because of verification. Lastly, these modern technologies assure maximum privacy by ensuring the restriction of sending confidential information.

Conclusion

In this paper, I have introduced new modern technologies that can be able to prevent fraudulent acts by preserving the privacy of the information. These modern technologies are differential privacy, federated analysis, homomorphic encryption, zero-knowledge proofs, and secure multiparty computation. Organizations, servicing companies, and institutions, most especially financial institutions like banks, can adopt the use of these modern technologies due to their great benefits in terms of data protection and preservation during the circulation of information. I have also observed some of the importance of these modern technologies that can improve privacy greatly and at the same prevent fraud.

References

1. A. El Orche, M. Bahaj, and S. A. Alhayat. 'Ontology Based on Electronic Payment Fraud Prevention'. Colloq. Inf. Sci. Technol. Cist, vol. 2018-Octob, pp. 143–148, 2018.
2. A. Tatiana. 'Fraud Prevention by Government Auditors'. Iber. Conf. Inf. Syst. Technol. Cist., 2017.
3. C. Guo, H. Wang, H. N. Dai, S. Cheng, and T. Wang. 'Fraud Risk Monitoring System for E-Banking Transactions'. Proc. - IEEE 16th Int. Conf. Dependable, Auton. Secur. Comput. IEEE 16th Int. Conf. Pervasive Intell. Comput. IEEE 4th Int. Conf. Big Data Intell. Comput. IEEE 3, pp. 106–113, 2018.
4. E. Caldeira, G. Brandao, and A. C. M. Pereira. 'Fraud Analysis and Prevention in E-Commerce Transactions'. Proc. - 9th Lat. Am. Web Congr. LA-WEB 2014, no. 2004, pp. 42–49, 2014.
5. E. I. Tarmazakov and D. S. Silnov. 'Modern Approaches to Prevent Fraud in Mobile Communications Networks'. Proc. 2018 IEEE Conf. Russ. Young Res. Electr. Electron. Eng. ElConRus 2018, vol. 2018-Janua, pp. 379–381, 2018.
6. 'Five New Technologies that Can Prevent Everything from Fraud to Future Financial Shocks'. Press releases| World Economic Forum'. [Online]. Available: https://www.weforum.org/press/2019/09/five-new-technologies-that-can-prevent-everything-from-fraud-to-future-financial-shocks/ [Accessed: 21-Dec-2019].
7. K. Modi and R. Dayma. 'Real-Time Fraud Detection in Credit Card Transactions'. Data Sci. Warsaw, 2017.
8. M. A. R. Vol. '& ACCOUNTING'. Vol. 16, no. 2, 2017.
9. N. Balasupramanian, B. G. Ephrem, and I. S. Al-Barwani. 'User Pattern Based Online Fraud Detection and Prevention Using Big Data Analytics and Self Organizing Maps'. Int. Conf. Intell. Comput. Instrum. Control Technol. ICICICT 2017, vol. 2018-Janua, pp. 691–694, 2018.
10. O. Avdeyuk, D. Kozlov, L. Druzhinina, and I. Tarasova. 'Fraud Prevention in the System of Electronic Payments on the Basis of POS-Networks Security Monitoring'. Proc. 2017 10th Int. Conf. Manag. Large-Scale Syst. Dev. MLSD 2017, 2017.
11. P. Sarda, M. J. M. Chowdhury, A. Colman, M. A. Kabir, and J. Han. 'Blockchain for Fraud Prevention: A Work-History Fraud Prevention System'. Proc. - 17th IEEE Int. Conf. Trust. Secur. Priv. Comput. Commun. 12th IEEE Int. Conf. Big Data Sci. Eng. Trust. 2018, pp. 1858–1863, 2018.
12. UK Finance. 'Fraud The Facts 2019'. Pp. 1–53, 2019.