Mining Your Transaction Data For Gold

This is a guest post by Brent M. Eastwood, Founder of GovBrain, a software company that automatically predicts financial market prices based on government and political data. Your Fintech firm is thriving. You are satisfying customer needs by properly tracking consumer habits, securing their bank account and credit data, and offering them efficient transactions. But you have found that optimizing your customers’ experience creates enormous amounts of transaction data. It is the age of big data and data science – that you’ve heard over and over – so you’re confident your firm’s data is valuable in some way. Taking advantage of machine-generated data can offer many advantages, even to the smallest of startups. These tasks provide the ability to solve tricky business problems and successfully navigate many different Fintech sectors. The acquisition and analysis of transaction data could lead to more customers and improve the customer experience to boost your bottom line. Using supervised machine learning techniques and natural language processing of unstructured transaction descriptions, one can predict future consumer behavior based on past transaction data. According to machine learning expert, Louis Dorard, firms can use “historical customer data to map snapshots of customers taken at a given point in time.” This type of mapping can reveal how your customers will react in different situations in order to keep them happy and prevent them from leaving. Great you say, that sounds interesting, but how do I do it? The software development efforts for machine learning will probably be expensive, time consuming and technically challenging, right? Not to fear. One way to start a new data science program is to use R programming. The open source R Project is a language for statistical analysis and data visualization with extensive support, resources, and user groups. It’s free and runs on Linux, Windows, and MacOS. One could also use H2O, an open source math and machine learning engine for big data that brings distribution and parallelism to powerful algorithms which keeps the widely used languages of R and JSON as an API. If you are interested in learning more about how your transaction data with improved merchant and geographic tagging can be used for new business opportunities, contact John Bird at