Evolving Paradigms: How Machine Learning Is Transforming the Financial Services Industry

Published on August 7, 2018 by Mayur Bhakre

The financial services industry is always one of the first to be directly affected by a change in technology. What computers did in the ’90s and the internet in the past decade, artificial intelligence (AI) is doing now. Machine learning (ML), a subset of the AI universe, has been making ripples in the delivery of financial services, with market spending on ML initiatives forecast to reach USD47bn by 2020.

An algorithm-based approach, ML intuitively learns from data rather than relying on a rules-based software program. This makes it more organic and flexible in its applications. Although an old concept, its utility is expanding in line with advances in computational capabilities.

The extent and scope of its successful implementation depend on identifying the specific suite of problems it is capable of addressing best. Typical characteristics could include but are not limited to

• Generating a prediction, not a specific answer

• Focusing on problems that can be conceptualized with a comprehensive set of example-based and continuous data

• Predicting trends that will not be materially affected by external information not considered in the existing data stream

ML applications are widespread. They can range from detecting ever-evolving forms of fraud, automating repetitive back-end processes, ensuring compliance with regulatory matters, and developing customer-centric and personalized products to replacing manual and unproductive work. ML applications yield monetary benefits too. According to industry estimates, some European banks utilizing ML have recorded a 10% increase in sales of new products, saved 20% in capital expenditure, enhanced cash collection by 20%, and reported a 20% decline in the churn rate.

ML applications in the financial services industry include the following

• Fraud prevention: Employing ML-based advanced analytics to examine large repositories of data pertaining to fraud cases can help to identify patterns and signs of potential fraud, enabling the service provider to take diligent steps to prevent and minimize investigation time by a large extent, while maintaining accuracy. MasterCard has employed an ML-based system that provides real-time assessment of transactional behavior to distinguish genuine transactions from fraudulent ones and to limit the frequency of false positives.

• Credit scorecards and risk analysis: ML can play a pivotal role by constructing efficient credit risk assessment models, eliminating errors introduced by human decisions. The advent of digital and peer-to-peer lending has further accentuated the applications of advanced analytics and ML in predicting risk in real time and with utmost accuracy.

  • JP Morgan Chase has developed a proprietary ML algorithm called Contract Intelligence or COIN that can analyze multiple documents at once and extract relevant information in a few seconds, enabling the bank to process 12,000 credit agreements in several seconds versus in 360,000 man-hours previously

• Customer servicing: The disruption introduced by fintech startups is forcing major organizations to revive their customer engagement strategies to include more interactive and dynamic relationship management. Firms across the globe are incrementally adopting ML- and AI-based tools such as robotic process automation and natural language processing to aid customer management.

  • Bank of America’s virtual assistant, Erica, leverages cognitive messaging and predictive analytics to provide financial advice to its customers. Netflix’s ML-driven personalization and recommendation system saves it more than USD1bn a year

• Others: Regulatory compliance, surveillance, data quality assessment, and optimization of trading execution by hedge funds and broker dealers. For example, the UK’s Financial Conduct Authority (FCA) is actively involved in understanding the opportunity AI and ML afford in enforcing regulatory compliance.

Although in an early stage of development, ML is increasingly becoming crucial for financial services organizations to remain competitive. Enterprise investments in ML at the global level are expected to double over the next three years. After boarding the innovation bandwagon, majority of CIOs are either planning to use or are already using ML in their organizations.

Advancing the fintech revolution, ML applications range widely, able to transform the financial services value chain. Acuity Knowledge Partners expertise in this domain involves understanding and acting on market dynamics via industry studies, strategic analyses, and undertaking competitive benchmarking and other studies of strategic interest across the fintech and financial services technology spectrum. Our platform oriented offering Business Excellence and Automation Tool (BEAT) provides customizable, domain-specific technology and automation solutions for financial services industry.

Key sources:

https://www.techrepublic.com/article/understanding-the-differences-between-ai-machine-learning-and-deep-learning/

https://www.forbes.com/sites/louiscolumbus/2017/10/23/machine-learnings-greatest-potential-is-driving-revenue-in-the-enterprise/#3df4507541db

https://www.mckinsey.com/industries/high-tech/our-insights/an-executives-guide-to-machine-learning

https://www.cnbc.com/2017/07/13/uk-regulator-looking-to-use-a-i-machine-learning-to-enforce-financial-compliance.html

https://fca.org.uk/publication/documents/from-maps-to-apps.pdf

https://www.servicenow.com/content/dam/servicenow-assets/public/en-us/doc-type/resource-center/white-paper/wp-cio-global-pov.pdf

---

---