Seminar by Maya Ramanath
Harvesting, Searching and Ranking Knowledge on the Web
Maya Ramanath
Max-Plank Institute for Informatics, Saarbrucken
Date: Tuesday, September 1, 2009
Time: 10:00 AM
Venue: CS102. Abstract:
The Web has the potential to become the world's largest knowledge base. In order to unleash this potential, the wealth of information available on the Web first needs to be harvested -- that is, extracted and organized. There is a need for new querying techniques that are more expressive than those provided by standard keyword-based search engines. And, in order for the results to be useful to users, they need to be ranked.
In this talk, I will describe my work on harvesting, searching and ranking knowledge on the Web. First, I describe YAGO, a knowledge-base of facts derived from Wikipedia and Wordnet and highlight our recent efforts to enhance this knowledge-base by making use of human computation techniques. Second, I will talk about our graph-pattern-based query language used to query the knowledge-base and a novel, language-model-based scoring model that we have developed to rank the results. I will conclude by highlighting avenues for future work.