Goran Oreški is an Assistant Professor at the University of Juraj Dobrila Pula and ai.Shuttle project leader at FIPU.Lab. His research interest lie in AI applications in banking and autonomous vehicles. He is the founder and CEO of GO Studio, the company implementing AI solutions in the banking industry.
Training data preprocessing for AI credit approval
The impact of artificial intelligence on the banking industry is increasing; the result is an ongoing automation trend of business processes where advanced AI models play the leading role. One of these processes is retail loan approval, where artificial intelligence handles the loan request instead of a personal banker.
For a model to replace a human in any business, especially in banking, many challenges must be addressed to ensure good generalization. In this lecture, I will talk about the quality of credit data used for training, specifically about new research on credit dataset balancing.