Alberto Artasanchez in a principal data scientist and the director of the DBG Artificial Intelligence Lab at Accenture with experience spanning over 25 years, in multiple industries and enterprise domains but with a focus on the financial industry. He has an extensive background in artificial intelligence and advanced algorithms. Mr. Artasanchez holds 9 AWS certifications including the Big Data Speciality certification. He is an AWS Ambassador and publishes frequently in a variety of Data Science blogs. He is frequently tapped as a speaker on topics ranging from Data Science, Big Data and Analytics, Underwriting optimization and Fraud Detection. He has a strong and extensive track record of designing and building end-to-end machine learning platforms at scale.
Artificial Intelligence Lab Initiative – Loan Default Prediction
- The business of lending is the core banking product.
- Created and enhanced Sagemaker models to augment and optimize the underwriting process.
- We have been able to achieve an average lift of 14% from baseline in net return.
- Getting the best results using Random Forest which still has explainability, a key requirement in the underwriting process.
- Models can also prescribe loan size and recommended terms and not just “lend/don’t lend” decisions.
Artificial Intelligence Lab Initiative – Fraud Detection
- Fraud is still a pervasive and significant problem in banking causing billions of dollars in losses on a yearly basis.
- Developed models to discern fraudulent credit card transactions using the features usually available in e-commerce transactions.
- Due to the relative scarcity of real-world data, created algorithms and processes to generate synthetic data.
- Developed knowledge graphs in TigerGraph to pinpoint suspicious behavior and facilitate the creation of SAR reports.
Alberto Artasanchez is also the author of “Artificial Intelligence with Python (Second Edition)”