Speaker/Affiliation: Jeff Dalton, Google
Date/Time: Tuesday, November 25, 2014
Location: UTA 5.522 (1616 Guadalupe St.)
* Catered lunch will be provided by Google: 12:30-1:15pm
* Colloquium Talk: 1:15-2:30pm
* Open meeting with students: 2:30-3:30
* 2nd, different talk in IR class (UTA 1.208): 3:45-4:45
Host: Matt Lease
Talk Title: From Text to Knowledge: Inferring and Exploiting Entity Structure in Large Collections
Large unstructured text collections contain a wealth of diverse information. I will present research that infers and exploits the structure in text by connecting it to the real-world using entities contained in large scale knowledge bases, such as Wikipedia and Freebase. Starting from unstructured text I will present information extraction techniques for inferring rich entity-based representations linked to external knowledge resources. I will examine the challenges of performing extraction on a large collection of scanned historical books, part of the Million Book Project. The talk will include research on extracting historical entities, linking entities to large-scale knowledge bases, and leveraging entity relationships to automatically construct knowledge bases. I will discuss the opportunities for exploiting this inferred entity structure for exploratory search applications. Finally, I will talk about ongoing work on automatic knowledge base construction from the web and other sources at Google.
Jeff Dalton completed his PhD at the University of Massachusetts Amherst with James Allan in the Center for Intelligent Information Retrieval. His research focuses on the intersection of Information Retrieval and Natural Language Processing. As a Software Engineer at Google, he investigates Knowledge Discovery in the Machine Intelligence research group. Before graduate school, he also worked on the Search Engine Development team at Globalspec, a vertical search engine for engineers. Outside of the office, Dalton appeared on Gordon Ramsay's "MasterChef" program on Fox in 2011 and can be found exploring farmer's markets on Saturday mornings.
7:15am to 8:30am