This talk will cover the applications of Machine Learning in UK Financial Services and in specific in credit risk. The talk will outline the research conducted by 4most into approaches used in our ML Scorecard Development Framework for identifying optimal hyper-parameters to build more accurate credit risk scorecards models together with some of the interpretability techniques that can be used to understand the resulting ML model globally and at case level. The talk will cover a business case highlighting the use of R as the main platform to develop such models.
Eduardo Contreras is currently principal consultant at 4most where he specialised in developing machine learning credit risk models. Before 4most he has previously worked in companies like EY, Lloyds Banking Group and Citigroup developing analytics solutions and credit risk models.