See below for a Business analytics seminar that might be of interest to AI folks. -Xiaoli An Alternative Approach to Transparent Decision-Making with Deep Machine Learning Algorithms in Mental Health Autism spectrum disorder (ASD) is a developmental disorder that affects 1 in 36 children in the US (1). The economic cost is estimated to be $66 billion per year in the US, from medical care and lost parental productivity (2). Early diagnosis is crucial since it allows for early treatment and the best long-term outcome. However, identifying children at an early age is challenging due to the limited ale mental health expertise, long wait lists, and misdiagnosis. Automated diagnosing and early referrals through machine learning (ML) can be of significant help. However, there are significant challenges to be overcome. Even with this growing patient population, this is a low-resource domain. The datasets are small, expensive, and difficult to create. The data itself is extremely diverse and training ML models for high performance is difficult. Transparency of models and support tools is highly valued. Finally, transfer learning is essential when applying models to data from different clinicians, clinics, hospitals, and populations. We are creating deep ML algorithms that rely on free text data in electronic health records (EHR) to support non-expert clinicians in identifying children at high risk for ASD. I will show early and recent results of our algorithms (using LSTM, GRU, and BERT models), highlight obstacles and how we are attempting to overcome them: synthetic and parental data to increase data size, transfer learning between datasets, relying on redundancy to increase performance, and avoiding binary black box labeling to provide transparency. [A person in a black suit Description automatically generated] Gondy Leroy, Ph.D., is Professor of MIS, Associate Dean for Research at the University of Arizona's Eller College of Management, and Research Director for the Center for Management Innovations in Healthcare. Her research focuses on natural language processing (NLP) and machine learning (ML) as they are part of artificial intelligence (AI). She has won grants from NIH (NLM, NIMH), AHRQ, NSF, Microsoft Research, and several foundations. She earned a combined BS and MS (1996) in cognitive, and experimental psychology from the Catholic University of Leuven (1996) in Belgium and a MS (1999) and PhD (2003) in management information systems from the University of Arizona. She serves on the editorial board of the Journal of Healthcare Informatics Research, Journal of Database Management, International Journal of Social and Organizational Dynamics in IT, Health Systems, Journal of Business Analytics, and co-chairs several special issues, sessions, tracks, workshops, and conferences focusing on design science and healthcare IT. She serves frequently on NIH grant review panels. She is the author of the book "Designing User Studies in Informatics (Springer, 2011). Finally, she actively contributes to increasing diversity and inclusion in computing through mentoring students early in their academic careers starting in high school and through graduate school.