Journal of Operational Risk

The use of business intelligence and predictive analytics in detecting and managing occupational fraud in Nigerian banks

Chioma N. Nwafor, Obumneme Z. Nwafor and Chris Onalo

  • Analysis of the frauds and forgeries returns reported by Nigeria Banks show that 2008 and 2016 had the highest frequency of frauds. Both years recorded 16,783 and 16,751 cases of frauds respectively. In terms of the severity, 2009 recorded the highest loss at N37,179,898.00 followed by 2008 with a loss severity of N34,311,721.00.
  • Although the banking industry has witnessed a downward trajectory in loss severity during the past four years, the increasing frequency of fraud occurrence suggests the existence of a ‘fraud path’- i.e. loophole in the detection and prevention systems of banks.
  • Despite the fidelity bond insurance cover taken by deposit money banks to address the challenges of occupational frauds committed by staff, and the use of manual transactions monitoring approaches, available evidence suggests rising cases of occupational frauds in the industry.
  • The current study proposes a two-fold approach to tackling occupational frauds within the Nigeria banking landscape- the use of data mining/business intelligence (outlier detection system) which aims to learn from historical transaction information in order to retrieve patterns that allow differentiating between normal and abnormal activities and predictive analytics techniques along with continuous monitoring of transactional data.

The problem of occupational fraud is one of the most wide-reaching operational risk event types in the Nigerian banking system. This event type spans many departments, roles, processes and systems and causes significant financial and reputational damage to banks. As a result, fraud presents banks with a real challenge in terms of knowing where to start. One of the main aims of this paper is to use stochastic probability models to predict aggregate fraud severity and fraud frequency within the Nigerian banking sector using historical data. Another objective is to describe how banks can develop and deploy business intelligence (BI) outlier-based detection models to recognize internal fraudulent activities. As the volume of transaction data grows and the industry focuses more closely on fraud detection, BI has evolved to provide proactive, real-time insights into fraudulent behaviors and activities. We discuss the fraud analytic development process, since it is a central issue in real application domains.

Sorry, our subscription options are not loading right now

Please try again later. Get in touch with our customer services team if this issue persists.

New to View our subscription options

You need to sign in to use this feature. If you don’t have a account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here