Data Analytics, Machine Learning (ML) and Artificial Intelligence (AI) - when put together – are building a powerful platform for advanced statistical analysis of structured and unstructured data.
This will provide the solution for the problem of identifying statistically important relationships among alternative and traditional time series data sets which are the foundation of meaningful trading and investment decisions.
Says Faisal Husain, Co-founder and CEO, Synechron,”We recently released Synechron’s AI Data Science Accelerators for financial services firms which apply AI to financial services business problems. The Accelerators use correlation and causation analysis to solve complex business challenges by discovering meaningful relationships between events that impact one another and cause a future event to happen.”
Banks using data to understand consumer behaviour is not a new occurrence. As Anand Subramaniam, Head, Artificial Intelligence Practice, Aspire Systems says, “What banks are currently trying to achieve is that, they are trying to obtain data and use machine learning models which when fed with data can provide insights into a particular customer’s behaviour. Data Analysis comes in the initial phase when banks are trying to make sense of several amounts of unstructured or structured data. Using an AI/ML model enables banks to create a forecasting model for a particular customer which also enables them to create solutions for futuristic problems which the customer might face, through in-depth predictive analysis.”
While most banking industry technology experts consider this to be the turning point for the industry, they recommend identifying areas that will benefit the most here. Says Sudhir Babu, Vice President and Head of Product Engineering at i-exceed, “we expect the usage of AI and ML in data analytics to play a transformative role in the banking sector. While their application is still in its early stage, two areas are likely to benefit the most from the first wave of AI and ML adoption. The first area is in the way banking is personalized. AI will help banks to gather unprecedented number of data points on how their customers are using various touch points and use them to customize the bank’s applications, products and services to better meet the customer expectations. The second area is in security and fraud prevention. There are ML models that help in creating a behaviour profile of each user based on the way the user interacts. After that, new transactions can be compared with the base profile and an additional authentication mechanism to prevent frauds can be introduced in case of differences.”
Until quite recently the complexity of the query required an agent to help and clarify products and services to customers in the sector, but it is no longer a novelty for them to routinely interact with AI driven bots in their banking experiences. “As these bots become more sophisticated, the nature of the agent’s role will transform into what the industry calls a super-agent, sophisticated bankers able to handle complex interactions while leaving the mundane to an algorithm,” says Chris Arnold (The Data Whisperer) - Senior Vice President, heading Product, Analysis, & Modelling at Enterprise Global Services, Wells Fargo. “Historically, when trying to understand customer feedback, banks would sample customer communications and have people write reports trying to summarize the feedback. While directionally helpful, the approach cannot be quantified. Natural Language Processing (NLP) can work on a 100% sample and decode the unstructured data into the key messages, providing a useful tool in addition to the traditional qualitative techniques.”
While these algorithms are most visible due to the direct interaction with customers, Machine Learning plays an increasingly important role in the back-end, he adds, from predicting systems outages to fraud detection and identifying terrorist financing activities.
Interestingly, Faisal sees a use for ML beyond merely customer connect. ”Traditional buy-side and sell-side research analysts will use tools backed by Data Science and AI/ML capabilities to study real-time market sentiments. This also will help to implement Granger Causality modelling efficiently into capital markets research to understand the impact that market events have on stock prices,” he says.
But how do experts see this area as evolving? Chris Arnold comments, “I’d describe the extent of analytics as lumpy and quickly evolving. India is ground zero for the world’s analysts across all industries. Estimates peg the number of analysts working in BFSI in India at over 50,000. As I meet with universities around the country, professors tell me that more and more of their students are headed for analytic jobs in BFSI. Many of my academic friends in statistics and economics feel they are coming out of the long shadow cast by their engineering colleagues! Clearly, this is driven by the usual suspects—large international players looking to take advantage of India’s dynamic population and world class educational system. However, leading banks and mobile payment gateways have become central pillars of the ecosystem and have fully embraced the journey to analytic driven businesses.
Conclusions
Banks have been using data to understand customer behaviour and to derive insights, so as to provide better experiences and enhance the customer satisfaction. Overall data analytics, AI and ML have a lot to offer for banks with respect to services such as customer segmentation into taste groups and excellent product recommendations and efficient and intelligent search engines. “It will be interesting to see how banks take up the available technology and how they leverage it in 2019,” concludes Anand Subramaniam.
Discover the latest Business News, Sensex, and Nifty updates. Obtain Personal Finance insights, tax queries, and expert opinions on Moneycontrol or download the Moneycontrol App to stay updated!
