*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Summary Sentence: Linking Human and Machine Discovery for Interactive Streaming Analytics at Scale
Full Summary: No summary paragraph submitted.
William A. PikeSpeaker:
William A. Pike
Director, Computational and Statistical Analytics Division at Pacific Northwest National Laboratory
Title:
Linking Human and Machine Discovery for Interactive Streaming Analytics at Scale
Abstract:
Contemporary analysis problems frequently derive from high-throughput streaming data sources, whether from scientific instrumentation, distributed sensor networks, or web-scale systems. Frequently, in-stream analysis methods for such data are based on rule sets, alerting thresholds, or incremental algorithms that require relatively little human input; when human-in-the-loop analysis is required, such exploratory discovery is often performed by discretizing the problem into static subsets that can be investigated offline. However, it is increasingly desirable to create interactive, visual analysis environments that can help make sense of dynamic phenomena as they occur, especially in domains where human judgment is critical like scientific discovery or threat detection, and where intervention in the dynamic phenomena requires real-time, rather than forensic, analysis. This talk discusses new research, open challenges, and collaboration opportunities in analysis in motion, an approach for bridging machine-driven and human-driven discovery in massive streams. This approach transforms the feedback loop in joint human and machine reasoning by capturing human background knowledge through visual interaction, automating the construction of human-useful hypotheses from live data, and steering models and data collectors in response to evolving knowledge.
Bio:
Bill Pike is the Director of the Computational and Statistical Analytics Division at the Pacific Northwest National Laboratory. His division leads research in data science, visualization and human computer- interaction, cyber security, and software engineering. He also leads the Analysis in Motion initiative, a five-year internal Laboratory investment to develop transformative analytics capabilities for large-scale, high-throughput data by blending human and machine reasoning capabilities in new ways. Dr. Pike has led research and development programs that have resulted in new information analysis methods in applications as varied as threat discovery, energy reliability, disaster response, cyber security, and privacy protection. He has also led the deployment of these capabilities to operational use in government and industry. He develops programs and advises government partners on visual analysis systems, distributed data-driven decision making, and human-computer interaction. He has also led the development of long-term visions for the future of information work for the US Government. He previously served as the Technical Group Manager for Visual Analytics at PNNL, the R&D coordinator for the National Visualization and Analytics Center, and the General Chair of IEEE VisWeek and the IEEE Conference on Visual Analytics Science and Technology. He holds a Ph.D. from Penn State.
