Big Data for Fraud analytics

In past, the auditors used traditional tools for comparison & analysis to detect chances of happening of any fraud. However, now with an increase in the volume, variety & velocity of data, these conventional tools lack the ability to process the data for the desired results. Now, the business data is not restricted to only the internal data generated for the business transaction but also includes the external data, like social media, system security logs, network logs, etc. With the varied data sources & the speed with which this huge volume of data is processed, the old analysis tool seems to be obsolete. Thus, the big data emerged with the ability to get the desired result. In simpler term, “Big data” is an evolution of long-standing information disciplines. It expands the range of data sources, the amount of data that can be processed and the speed with which answers can be derived.

According to ISACA, “Big data” refers to large, quickly growing or varied types of information i.e. high volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization.

Performance improvement of any business process is a primary objective of an organization but it would be more advantageous when there is an increase in quality of processed output. Big data separates itself from other information management approaches by providing quality in processed output. One of the major benefits of Big data analytic is prevention of frauds.

Companies around the world lose US$3.5 trillion to fraud each year, according to a report by the Association of Certified Fraud Examiners. Keeping this is view, it becomes evident for each organization to invest in data analytic for fraud management. Handling fraud manually is not only too expensive but also chances of existence of high value frauds go undetected. The growth in use of unstructured data leaves a lot of room for fraud. Thus, in the present context, it become important to analyze real time structured & unstructured, internal & external data to detect & prevent frauds. Insurance companies where the happening of frauds is too high, has been using this big data analytic for fraud management.

Big data analytics plays an important role in data integration by associating information within organization. Big data analytics goes a long way in detecting & managing fraud proactively in business lifecycle. It helps in plummeting the overall cost of fraud detection and improving the Return On Investment. Big data analytics can be used in any business process for fraud detection & prevention be it a procurement cycle, revenue cycle, expense cycle or for management analysis. It associates multiple applications and data sources together to enhance this data & permits the data to be probed and visualizes for further analysis, providing deeper insights and answers.

In this paper, we will focus on travel & expense process. This is a business process where the chances occurrence of fraud is high as it involves the foreign currency. Big data analytic approach has few steps in analyzing fraud in the business process which may range from analysis of social media, doing a predictive analysis, performing the SWOT analysis, data mining, integrating management system to control social media, performing analysis of customer relationship management, statistics, and natural language processing, etc.

Business travel and related expenses in any organization plays a critical role in expense management. A large chunk of corporate budget is allocated to travel & expense process. Thus, it becomes so vital to consider all possible value additions to the travel & expense process improvement. For this value addition to the travel & expense process, big data analytic may consider the below few steps like:

1. Understand how business travel expense effects the entire organization. The first step in improving the travel & expense process is to understand from where travel expense emerges. Who’s traveling? How often are the travels undertaken? How much is the quantum/amount of travel Expenses? The more information gathered, the better analytics can be performed.
2. Address the weakest activity in the full lifecycle of travel & expense process from travel-booking, expense workflow, creation of travel request, submission of application for travel, approval and validation of travel application, submission of travel reports & documents, reimbursement of travel Expense for back-end analytics.
3. Big data analytics revolves around three things, data, reporting & analytics. Value addition to the travel & expense process can happen by analyzing the data & the travel reports & negotiating for the best contracts for travels.
4. Calculate the exact return on travel investments.
5. Understanding the travel supplier base.
6. Identifying & predicting, instances of non-compliance .This in return will help in building long term forecast for financial expenses based on current data.
7. An effective travel and expense management that simplifies travel-booking and ensures a holistic series of expense management processes can drive significant cost reductions over the long run and ensure that key stakeholders have true acumen for long-term planning and development.

Thus, to conclude, the need for new tools for analytics, availability of huge data storage capability led the emergence of Big data analytics. And Big data analytics can be used for performance improvement of any business process by providing high quality processed data which is fraud free.

Contributor: Anand Prakash Jangid with Shefalika Sahu from Quadrisk Advisors