Explainable AI Using h2o AutoML and Robustness Check in Credit Risk Management

The paper proposes an eXplainable Artificial Intelligence model that can be utilized in credit risk the board and, specifically, in estimating the dangers that emerge when credit is acquired utilizing shared loaning stages. The model applies connection organizations to Shapley esteems with the goal that Computerized reasoning expectations are gathered by the likeness in the hidden clarifications. The exact examination of 15,000 little and medium organizations requesting credit uncovers that both hazardous and not dangerous borrowers can be gathered by a bunch of comparative monetary qualities, which can be utilized to make sense of their FICO rating and, in this manner, to foresee their future way of behaving.