The Convergence of Data Science and Financial Services

Oct 15, 2019

An abundance of data and rapid decision making characterizes the financial services industry sector making it a ripe ground for the use of data science technologies. Finance firms are relying on these technologies more and more, to the extent that job listings in the financial industry, requiring data science skills, have increased 60% in the past year. Two uses stand out for data science in financial services: improving decision making and automation. Decision making is improved, as machine learning techniques are used to make increasingly accurate predictions. Techniques like natural language processing (NLP) are used to automate repetitive tasks that deal with large amounts of text.

Data science is being applied to many tasks to improve risk management. Detecting fraud has become easier to do with the help of advanced machine learning and artificial intelligence (AI) algorithms. Data science adds value to fraud detection by allowing for sophisticated, large scale, and real time pattern detection that can adapt with current trends. This advantage is shown through the number of financial institutions trusting data science technologies. For example, a recent study has shown that about 73% of financial institutions with over $110B in assets are currently using AI for payment fraud detection. An additional application of data science is in the scoring of credit risk. Machine learning techniques allow for new larger data sets to be incorporated into credit scoring models, in some cases up to 10x more data. This allows for more accurate risk predictions and extends the potential pool of loan applicants to those who have not been able to build up traditional credit scores.

NLP allows computers to process and analyze large amounts of text and speech data. The application of these techniques has been saving financial services firms money and time by automating laborious tasks. For example, JPMorgan Chase developed a text mining solution that can analyze commercial loan contracts and saves 360,000 hours per year in staff time. NLP finds many use cases in customer service. For example, many customer facing tasks are being handled by chat bots, as opposed to human staff. USAA, Bank of America, Capital One and other financial services firms have all implemented chat bots through which the customer can get information and make decisions about their accounts. Chat bots utilize NLP to understand customers’ requests and provide answers that feel natural in the conversation. By 2020, Gartner predicts that customers will manage 85% of their relationship with companies, without interacting with a human.

In financial services, it is important to keep an edge. Right now, those with data science capabilities are pushing the edge farther and farther from those who lack those capabilities. The insights gained from mining massive datasets allows for more profit and less risk. Utilizing data science to create automated processes cuts cost and speeds up operations. In order to stay competitive, it is imperative that financial managers learn to unlock the value of their data.