Technology is developing at an unimaginable speed and the rapid change and development is posing the entities with new risks and threats that might make the entities more vulnerable to fraudulent activities and Banking sector is more prone to these frauds as the systems are directly or indirectly approachable by a large number of stakeholders. While development in Technology is inevitable, it is also important to keep a check on the loopholes and probable risk areas to ensure these are identified at an early stage to avoid massive losses.
Banking fraud is a crime of intentionally attempting to execute a scheme to defraud a financial institution and obtain their properties through fake representations. The banking industry has gone through various changes post liberalization, though the industry is well regulated, but it suffers from its inherent weaknesses. Increase in the number and types of frauds in this sector has posed a challenge to the industry. Although, banking frauds in India have often been treated as the cost of doing business, the complexity and cost of it have increased manifold.
RBI, the regulator of banks in India, defines fraud as “A deliberate act of omission or commission by any person, carried out in the course of a banking transaction or in the books of accounts maintained manually or under computer system in banks, resulting into wrongful gain to any person for a temporary period or otherwise, with or without any monetary loss to the bank”. The rising NPAs and the increasing involvement of the senior management in the frauds have caused the amount of fraud to go manifold. The robustness of a country’s banking and the financial system is a prerequisite for economic growth and development.
Early detection of fraud plays a major role in minimising losses that could have occurred if the act of fraud continued to be undetected. Statistical & Data analysis is the most effective tool that can not only ensure that the required controls are in place in the system but are also effective round the year, thus enabling early detection of fraudulent activities. One of the most widely used tool of Statistical and Data analysis is Artificial Intelligence. Both these tools have been explained below:
Statistical & Data Analysis Techniques: The statistical & data analysis techniques are helpful in exploration, comprehensions and decision-making. A lot of software-based tools are being used by financial institutions to detect fraud. These software tools may be available in various categories of statistical & data analysis techniques detailed as below-
|1.||Data Pre-processing techniques||This technique is being used for detection, validation, error correction and filling up of missing or incorrect data. Hence, it helps in Data cleaning by filling in missing value, smoothen noisy data, identify or remove outliers and resolve inconsistencies, Data integration by using multiple databases, datacubes or files and Data transformation by using normalization and aggregation techniques.|
|2.||Statistical parameters||Probability distributions, Performance Metrics, etc.
|3.||Time dependent data||Used for insurance, recurring deposit, loan transactions|
|4.||Time series Analysis
|It is used to predict future values based on previously observed values. It comprises methods for analyzing time series data in order to extract meaningful statistics and other characteristics of the data.|
|The task of grouping a set of objects in such a way that objects in the same group (i.e., cluster) are similar (common traits or properties) to each other than to those in other groups or clusters. It is the main task of exploratory data mining. It is a common technique for statistical data analysis used in many fields of fraud detection like pattern recognition, information retrieval and machine learning.|
|6.||Matching Algorithms||It is used to detect anomalies in transactions or users’ behavior compared to previously known models and profiles. This is very useful in credit card fraud detection.
Artificial Intelligence: AI is one of the most popular Statistical & data Analysis Technique. AI detects fraud real time and thus improves processing speed, efficiency and accuracy, banks can, therefore, approve transactions faster with less false positive.
AI techniques used for fraud management generally includes-
|1.||Data Mining||Data mining is the process of discovering patterns in large data sets involving methods at the intersection of machine learning, statistics and database systems. So, data mining is to classify, cluster and segment the data and automatically find associations and rules in the data that may highlight pattern of fraudulent transactions.|
|2.||Expert System||Knowledge based Expert system is used to develop a software that stores all the human expertise (after being input by the human expert himself) and then using stored human intelligence to detect fraud.|
|3.||Machine learning and pattern Recognition||Machine learning is closely related to computational statistics which also focuses on prediction making through the use of information technologies. Machine learning can help also be unsupervised and be used to learn and establish baseline behavioral profiles for various entities and then find meaningful anomalies related to fraud or any other transactions. Pattern recognition is a branch of machine learning that focuses on the recognition of the patterns and irregularities in data. Pattern recognition systems are in many cases, trained from labeled training data (i.e., supervised learning), but when no labeled data are available, other algorithms can be used to discover previously unknown patterns (i.e., unsupervised learning).|
|4.||Neural Networks||Neural network-based fraud detection is based totally on the human brain working principal. Neural network technology has made a computer system capable of think. The inherent nature of neural networks is the ability to learn is being able to capture and represent complex input/output relationships.|
The banks should undertake the following pre-employment checks at the minimum to mitigate frauds:
- Background checks through professional agencies including police verification
- Employment checks with the previous employer
- Conduct regular fraud risk assessments. A good fraud risk assessment generally answers the following three questions:
- Am I aware of the fraud scenarios in my immediate environment?
- Am I aware how a potential fraudster overrides or circumvent existing systems and controls?
- How is the effectiveness of controls monitored?
A team of specialists can be formed to collect information on the latest fraud schemes and test existing controls for vulnerability.
- Invest in gathering of Intelligence.
“Mystery Shopping” is an important element of fraud vulnerability assessment. This will enable banks to test the efficacy of controls to existing and new fraud scenarios and to identify collusion, if any, which could result in the circumvention of controls.
- Use dedicated Forensic tools during an investigation process.
Forensic tools can be used to navigate IT systems for evidence of malfeasance, such as information deletion, policy violations and unauthorized access. A wealth of information can be recovered from computers, including active, deleted, hidden, lost or encrypted files or file fragments, which can be presented in a court of law. These include tools of forensic imaging, electronic discovery, data anomaly detection, and record management, which can help banks and their legal counsels in handling and analyzing large and complex data issues to help support their cases.
A constant eye on the probable fraudulent activities has become an inseparable part of any banking organization and the earlier it is detected or even better prevented, the lesser the damage will be. Strong, effective and well-run banking organizations exist because the management tends to take proactive steps to anticipate probable risk and take preventive action to mitigate these risks and avoid undesirable fraudulent actions. In the current times of continuous and quick technological changes it is inevitable to ensure continuous assessment of fraud exposures.