Posted by Bethany Frank on 03 Oct 2016
Machine learning is a branch of cognitive computing that enables computers to learn from data with minimal human intervention. Recent advances in technology have enabled financial institutions to explore the applications of machine learning techniques in areas like customer service, personal finance and wealth management, and fraud and risk management.
Natural Language Processing (NLP) is a subset of machine learning that enables systems to understand language. Banks are using the technology to build automated voice systems and chatbots to help customers make payments, manage their accounts, and find answers to general inquiries without the help of a live representative.
Chatbots and similar technologies enable banks to deliver an enhanced customer experience. For instance, if a customer tends to pay the minimum amount due as late as possible, it might indicate he or she is facing a cash-flow crunch. This could be a lead for a financial institution to offer the customer a loan. Payment trends and other behavioral aspects are valuable inputs to determine what kinds of loans customers are eligible for.
Augmenting traditional customer service with AI-based digital assistants can help customers resolve inquiries faster and enable banks to have more efficient support teams. A system could learn from customer and agent behavior to eventually resolve certain kinds of issues automatically, freeing up time for staff to address larger priorities.
Budget management apps armed with machine learning capabilities provide customers the benefit of highly targeted financial recommendations. The technology can help customers track spending on a minuscule level and access such information through the digital channels they spend their time on. It can also provide valuable data on every facet of consumer spending and behavior.
Machine learning helped one international bank identify struggling customers and better serve them. When the system detected unusual spikes in late-night credit card usage among certain customer segments combined with low savings rates, the bank was able to determine that these customers were all facing some level of financial stress. In order to reduce the risk of defaults, customers tagged by the system were automatically given credit limit increases and financial advice.
Improved technology would also make financial advisory services more accessible to larger segments of a bank’s customer base. Robot advisers and similar apps help take the guesswork out of money management, a skill many key demographics lack. Such services can help a bank’s customers establish solid financial standing without much manual effort.
Fraud and Risk Management
Financial institutions today deal with an unprecedented amount of consumer information that must be scanned for fraud. As transaction volumes increase and consumers adopt new ways to pay, banks will need to strengthen fraud and risk management protocols to handle new demand.
Companies like PayPal are already using machine learning to combat fraud. Through a combination of linear, neural network, and deep learning techniques, PayPal’s risk management engines are able to determine the level of risk associated with a customer within milliseconds. Similarly, banks can replace statistical risk management models with machine learning algorithms. Several banks already utilize programs that scan transactions for risk, move flagged transactions to a risk queue, and further investigate for suspicious activity automatically. Continuously expanding knowledge bases improve the ability of such programs to detect fraud over time.
Insights gathered by machine learning programs could provide financial institutions with actionable intelligence that serves as a foundation for various decisions. For example, a machine learning program could tap into various data sources to assign risk scores to customers applying for loans. Algorithms would be able to predict which customers are at risk for defaulting on a loan, enabling the bank to tailor services or term adjustments for each customer.
Machine learning can also help banks determine the creditworthiness of potential customers. By analyzing past spending behavior and patterns, a system could identify how much credit should be extended to a given customer. The technology would be especially useful in the case of new customers or those who lack a long credit history, i.e. millennials. Automating credit and risk scoring processes on a mass scale can help banks enhance their credit and risk scoring models across the board.
The applications of machine learning in the financial services world extend well beyond these few examples. Apt implementation of machine learning technologies can help banks enhance security, offer customers better digital service, and increase operational efficiency. For more on financial technology and innovation, visit our Resource Center.