Abstract: Organizations are investing in Big Data and AI, but the majority of these projects were predicted to fail. A study shows that one of the biggest obstacles is the lack of understanding how to use analytics to improve business. This paper presents Metis, a method for ensuring that business goals and problems are explicitly traceable to Machine Learning projects and potential (or hypothesized) complex problems can be properly validated before investing in costly solutions. Using this method, business goals are captured to provide context for hypothesizing business problems, which can be further refined into more detailed problems to identify features of data that are suitable for machine learning (ML). A supervised ML algorithm is then used to generate a prediction model that captures the underlying patterns and insights about the business problems in the data. A ML Explainability model is then used to extract from the prediction model the individual features and the degree of which contribute to the problems. The extracted feature contributions are then fed back to the goal-oriented problem model to validate the most important business problems. Our experiment results shows that Metis is able to detect the most influential problem root cause when it was not apparent through data analysis. This approach is illustrated using a real-world customer churn problem for a bank and a publicly available customer churn dataset.
Authors: Sam Supakkul (NCR Corporation, USA); Robert Sungsoo Ahn and Ronaldo Goncalves (University of Texas at Dallas, USA); Diana Villarreal (NCR Corporation, USA); Liping Zhao (University of Manchester, USA); Tom Hill and Lawrence Chung (University of Texas at Dallas, USA)
Email: ssupakkul@ieee.org, ronaldo.goncalves@utdallas.edu, diana.villarreal@ncr.com, liping.zhao@manchester.ac.uk, tom.hill.fellow@gmail.com, chung@utdallas.edu