When: May 27th, 2019 | 12:30 PM - 1:30 PM
Where: Room X836, ICICS/Computer Science Building (X wing), 2366 Main Mall
Please RSVP here: https://bit.ly/2VoofgN
Monitoring, interpretation, and analysis of large amounts of time-stamped, multivariate data are subtasks that are at the core of tasks such as the detection of malware in communication data, the integration of homeland security information from multiple sources, the management of chronic patients using clinical guidelines, the retrospective assessment of the quality of the application of such a guideline, and the learning of new knowledge from analyzing the data regarding repeating patterns of process-related actions, of measured data, and of meaningful abstractions derivable from these data.
I will briefly describe several conceptual and computational architectures developed over the past 20 years, mostly by my research teams at Stanford and Ben Gurion universities, for performance of these tasks, often exploiting domain knowledge, and will highlight the complex and interesting relationships amongst them. I will focus on the differences and similarities between interactive visual-exploration frameworks for single and multiple longitudinal records, and data-driven frameworks for temporal data mining, and how both can be used and even integrated for clustering, classification, and prediction. I will focus on the medical domain.
I will also talk about classification and prediction techniques that build on the capability to create meaningful interval-based abstractions from the time-oriented multivariate data, such as interval-based extensions to dynamic time warping (iDTW), the use of frequent time-interval relations patterns (TIRPs) discovered within the interval-based abstractions, as features, and Temporal Probabilistic Profiles (TPFs), which characterize entities by their TIRP distribution, which we are presenting in our coming KDD-2019 paper.
I will describe these topics as a progression, exemplified within the medical domain, from individual subject monitoring, diagnosis, and therapy, to multiple-subject aggregate analysis and research, and finally to the learning of new knowledge from large numbers of longitudinal records.
About the Speaker:
Professor Yuval Shahar holds advanced degrees in Medicine, Computer Science, and Medical Informatics. A former researcher and faculty member at Stanford University, he moved to Ben-Gurion University to found and head its Medical Informatics Research Center, and is the Josef Erteschik Chair of BGU’s Department of Software and Information Systems Engineering.
Light refreshements will be served.