Automatic Set Instance Extractionusing the Web.ppt

Language Technologies Institute, Carnegie Mellon University *Automatic Set Instance Extraction using the Web Richard C. Wang *Automatic Set Instance Extraction using the Web Automatic Set Instance Extractionusing the Web Richard C. Wang and William W. Cohen Language Technologies Institute Carnegie Mellon University Pittsburgh, PA 15213 USA Challenge Discovering set instances, or hyponyms, of any given semantic class name x is a hyponym of y if x is a (kind of) y * / 21 Automatic Set Instance Extraction using the Web “Failed Banks” “Bags” These are real examples from our system described in this paper “Hair Styles” * / 21 Automatic Set Instance Extraction using the Web Outline Background – SEAL Proposed Approach – ASIA Evaluation Conclusion * / 21 Automatic Set Instance Extraction using the Web Background – SEAL Set Expander for Any Language Wang Cohen, ICDM 2007 An example of set expansion Given an input query (seeds): { survivor, amazing race } The output answer is: { american idol, big brother, ... } A well-known SE system is Google Sets? /sets * / 21 Automatic Set Instance Extraction using the Web Background – SEAL Features Independent of human markup language Support seeds in English, Chinese, Japanese, Korean, ... Accept documents in HTML, XML, SGML, TeX, WikiML, … Does not require pre-annotated training data Utilize readily-available corpus: World Wide Web Research contributions Automatically construct wrappers for extracting candidate items Rank candidates using random walk * / 21 Automatic Set Instance Extraction using the Web SEAL’s Pipeline Canon Nikon Olympus Pentax Sony Kodak Minolta Panasonic Casio Leica Fuji Samsung … Fetcher: Download web pages containing all seeds Extractor: Construct wrappers for extracting candidate items Ranker: Rank candidate items using Random Walk Can you find common contexts that bracket every seed instance? I guess not! Let’s try our Extractor … Our Extractor finds maximally-long contexts that bracket at least one