Approximately one in 25 adults in the U.S. suffers from a severe mental illness such as depression or schizophrenia that interferes with life activities; this rises to one in five adults for mental illnesses across the board. Annual costs for schizophrenia treatment exceed $6.8B annually, while the most recent data for depression show over $40B in annual costs. Suicide is the tenth leading cause of death in the U.S., the second leading cause of death for people aged 15 to 34, and the number of suicides in the U.S. is more than double the number of homicides. If we can use technology to make progress on mental health, it will make a big difference. In this talk, I will discuss my work on natural language processing and machine learning methods in mental health. This includes technological development, but it also includes efforts to develop a broader community focused on these issues and technical infrastructure that allows researchers to work together more effectively on sensitive data.
Philip Resnik is Professor at University of Maryland, with joint appointments in the Department of Linguistics and the University of Maryland Institute for Advanced Computer Studies. Hiss current NLP research focuses on computational social science, particularly mental health: he was a co-founder of the CLPsych workshop series, a member of the Technology and Innovation Committee of the American Association of Suicidology, and his 2018-2019 sabbatical project, supported by an Amazon Machine Learning Research Award, involved building a secure mental health data enclave to facilitate shared research access to sensitive datasets.