-
Video
-
Table of contents
-
Video


- Description
- Transcript
- Discussion
About the talk
00:00 Intro
00:15 Disclosure
00:45 Main Questions
01:08 Methods
01:36 Multimodal Classification
01:58 Results
02:23 Subjective experience
02:39 Conclusion
About speakers
I am a CIHR Health System Impact Postdoctoral fellow in the Department of Computer Science and the Data Science Institute at the University of British Columbia and at the BC Centre for Disease Control. I work with Giuseppe Carenini, Raymond Ng, Naveed Janjua, Thalia Field, and Richard Lester. I received my Ph.D. from the Language Technologies Institute at Carnegie Mellon University. My advisor was Carolyn Rose, and my thesis committee members were Eduard Hovy, Louis-Philippe Morency, and Ekaterina Shutova. My work explores language use from a computational perspective to facilitate human communication. Using Machine Learning and Natural Language Processing (NLP) techniques, I discover patterns in text and build computational models of language use that are able to process large amounts of linguistic data. I investigate how people use language by bridging computer science and the disciplines of linguistics and psychology in the health care domain.
View the profileDr. Giuseppe Carenini, PhD, is a Professor in Computer Science and the Director of Master of Data Science program at UBC. His work on natural language processing and information visualization to support decision making has been published in over 130 peer-reviewed papers (including best paper at UMAP-14 and ACM-TiiS-14). Giuseppe served as Senior Area Chair at NAACL; Area Chair at ACL, NAACL, EMNLP; and as Program Chair at SigDial and IUI. In 2011, he published a co-authored book on “Methods for Mining and Summarizing Text Conversations”. Giuseppe was awarded a Google Research Award, an IBM CASCON Best Exhibit Award, and a Yahoo Faculty Research Award in 2007, 2010 and 2016.
View the profileDr. Conati is a Professor of Computer Science at the University of British Columbia, Vancouver, Canada. She received a M.Sc. in Computer Science at the University of Milan, as well as a M.Sc. and Ph.D. in Intelligent Systems at the University of Pittsburgh. Conati’s research is at the intersection of Artificial Intelligence (AI), Human Computer Interaction (HCI) and Cognitive Science, with the goal to create intelligent interactive systems that can capture relevant user’s properties (states, skills, needs) and personalize the interaction accordingly. Conati has over 100 peer-reviewed publications in this field and her research has received awards from a variety of venues, including UMUAI, the Journal of User Modeling and User Adapted Interaction (2002), the ACM International Conference on Intelligent User Interfaces (IUI 2007), the International Conference of User Modeling, Adaptation and Personalization (UMAP 2013, 2014), TiiS, ACM Transactions on Intelligent Interactive Systems (2014), and the International Conference on Intelligent Virtual Agents (IVA 2016). Dr. Conati is an associate editor for UMUAI, ACM TiiS, IEEE Transactions on Affective Computing, and the Journal of Artificial Intelligence in Education. She served as President of AAAC as well as Program or Conference Chair for several international conferences including UMAP, ACM IUI, and AI in Education.
View the profileDr. Thalia Field, an Associate Professor in the Division of Neurology, is a stroke neurologist and clinician-researcher with a focus on clinical trials. She holds the Sauder Family/Heart and Stroke Foundation Professorship of Stroke Research. She is currently leading a national study examining treatment strategies and prognosis of cerebral venous thrombosis, a rare cause of stroke primarily affecting younger women. She has a particular interest in process improvement in clinical trials, including working with patients to identify important outcomes, and integrating existing and emergent technology to enhance efficiency and engage under-represented populations. She completed her MD at Dalhousie University, followed by her neurology training at the University of British Columbia and stroke fellowship and epidemiology training at the University of British Columbia and the University of Calgary.
View the profileTom Soroski started with the CANARY team in September 2018 as a co-op research assistant while completing his undergraduate degree. Tom was part of the CANARY project from its inception; applying for ethical approval, helping set up the data collection protocol, and recruiting the first study participants. Tom graduated from the University of British Columbia in 2020 with a Bachelor’s of Science in Microbiology and Immunology, and returned to work on the project full-time as the study manager. Tom enjoys being part of CANARY project as it allows him to work closely with patients and their families, and to contribute to Alzheimer’s disease research.
View the profileSally Newton-Mason is a research assistant with a Bachelor’s of Science in Biology from UBC. Previously she has worked as a care-aid, and loves interacting with families in a clinical setting. CANARY is a very special project to her as it combines her passion for healthcare with her curiosities in artificial intelligence. She loves that the research aims to innovate and improve healthcare paradigms. In the future she is planning to pursue a Bachelors of Nursing to further her opportunities in the Healthcare and Research fields. In her free time, she loves singing, painting, and spending time with her beloved cat.
View the profileToday only presenting our results on learning basic classification of Alzheimer's disease is night writing and language on behalf of my Quarters at the University of British Columbia. Currently there are no disease-modifying treatment for Alzheimer's disease, which would be most effective in individuals who do not yet, have advance in a rather than reactive changes. However, current strategies for individuals who have preclinical diseases are efficient, and imprecise. And there is a need for cost-effective accurate and non-invasive screening, tools of the law, to identify individuals
with breaking, the coronavirus disease from a beacon candidate for disease-modifying treatment FiOS. As a preliminary step towards this Vision will investigate the use of eye, tracking data collected during a spontaneous speech, task to distinguish memory Clinic patient from healthy controls later for this task and here we are again. Whether I second base models can be combined with language-based approaches to increase classification performance. We recruited 72 Basin from a specialized memory clinic and 68 healthy controls from the community. Participants were
asked to describe the cookie that in mind while their speech and eye movements were recorded which to operate between one and one and a half minutes to go get the picture. But I'm movements have not been saving this contact so far. Remind me, the combination of the two modalities will investigate two different approaches in the early season of instant, your vector before it said into the classifier, the late-season approach from the eye and language teacher separately. And their predictions are averaged to generate a final prediction. Results of the great features are predicted in our
data set and inside statistically, outperformed the date line, which was based of demographics and medical history data is complementary to language a thing. This classification that has the best performance was achieved when combining both know that I'm using PlayStation of living 2.8 in area under the rock cover. Most participants from the setup comfortable or relax during the assessment, and we're willing to repeat that segment on a yearly basis. As a long-term spending his money pain,
treatment trials, In conclusion, we collected during a spontaneous, which task is able to discriminate between patients and have the controls while performing the Baseline. Best results are received in the last season of broads. Indicating that I gave data is complementary to language data for the specification. The assessment task is well tolerated, which is promising towards a non-invasive efficient, cost-effective screening tool for this is modifying payment prior. Thank you.
Buy this talk
Ticket
Interested in topic “Artificial Intelligence and Machine Learning”?
You might be interested in videos from this event
Similar talks
Buy this video
Conference Cast
With ConferenceCast.tv, you get access to our library of the world's best conference talks.
