ORNL Invention to Innovation Webinar 5: Visual Text Analytics
(EDITOR’S NOTE: This is the next in a series of webinars that Oak Ridge National Laboratory is hosting to provide overviews of various inventions and highlights of specific commercial opportunities approaching market readiness. Each webinar concludes with a roundtable session for Q&A and market feedback.)
TECHNOLOGY OPPORTUNITIES IN INFORMATION TECHNOLOGY, SECURITY AND DEFENSE PORTFOLIOS
Featured Technology: Interactive Visual Text Analytics for Situation Awareness of Social Media
September 9, 2014 at 2 p.m. EDT
Oak Ridge National Laboratory (ORNL) plays an important role in national and global security by virtue of its expertise in areas such as supercomputing and data analytics. Innovation in these areas is essential for protecting US citizens and advancing national and global security priorities. ORNL Researchers support this mission by applying their scientific expertise to address critical challenges in areas such as cybersecurity.
With more than 140 million active users, Twitter produces about 340 million, 140-character posts each day. Virtual communities form rapidly on such social media platforms, and occasionally these virtual communities can foster violence and unrest that is conveyed in the users’ language. Techniques for analyzing broad trends over these networks or reconstructing conversations within small groups have been demonstrated in recent years but have not supported near real-time analysis of these large streams of information. ORNL researchers have developed a system to discover and interactively explore these virtual networks; detect sentiment such as fear, anger, or sympathy; highlight change and trends; and discover patterns within the network movements, such as use of certain hashtags.
The ORNL-developed system continuously monitors an online feed of randomly sampled public posts from social media such as Twitter. Researchers augment the sampled stream with additional filtered queries for specific geographical areas of interest and/or keywords to increase the volume of relevant data. In parallel, another automated service monitors and archives RSS news feeds for an alternate source of event information. The combined efforts allow researchers to analyze social media activity on a global scale. Social media and other similar online streams of textual information represent a largely untapped resource for awareness and understanding of global events.
Presenters
David Sims, Senior Commercialization Manager Chad Steed, Research Scientist
