- This event has passed.
MIDAS Seminar Series Presents: Michigan Data Science Fellows: Elyas Sabeti and Patrick Park
September 16 @ 3:30 pm - 4:30 pm
Michigan League, Vandenberg Room
This is the first joint seminar presented by the Michigan Data Science Fellows and the kick-off of the 2019-2020 MIDAS seminar series.
Dr. Elyas Sabeti: Predictive Analytics in Healthcare: From Anomaly Detection to Development of Clinical Decision Support Systems
Abstract: A central question in the era of “big data” is what to do with the enormous amount of information. Predictive Analytics answer this question by providing a platform in which historical data is being learned for the best decision to be made about future. In healthcare, the ultimate goal of this process is to detect an anomaly and to develop diagnostic tools. In this talk, examples of Predictive Analytics using information-theoretic and machine learning approaches with applications in healthcare are provided.
Dr. Patrick Park: The Strength of Long-Range Ties
Abstract: Four decades of small-scale social network research has widely assumed that interpersonal connections bridging socially distant groups are weak, comprised of sporadic and emotionally distant relationships. However, in this study using data from eleven culturally diverse population-scale phone and Twitter communication networks on four continents, I find that previously unobservable long-range bridging ties can be nearly as strong as the socially proximate non-bridging ties that share a common friend. Evidence from content, temporal, and geographic analyses suggests that these strong connections spanning extraordinarily distant network communities are emotionally expressive, socially oriented relationships. The discovery of relationally strong, long-range bridging ties have implications for the speed and breadth of social contagion.