170 likes | 261 Views
SenseAble Search. Shailesh Kochhar and Adam Vogel CS 498CXZ, Spring 2006. Ambiguity. Words are ambiguous But they don’t have to be Otherwise we’d never understand each other. So now, when is…. A golf club not a golf club? A chair not a chair? A bill not a bill?. When….
E N D
SenseAble Search Shailesh Kochhar and Adam Vogel CS 498CXZ, Spring 2006
Ambiguity • Words are ambiguous • But they don’t have to be • Otherwise we’d never understand each other
A golf club not a golf club? • A chair not a chair? • A bill not a bill?
It allows you to rent a golf cart. • He or she calls a meeting to order. • People vote on it, or it comes with a duck
Ambiguity Detection • Tag sense of query terms in top documents • Examine the distribution of senses • Ambiguity = Large number of senses • Diverse distribution = More random • Measure randomness?
Ambiguity Resolution • Use entropy of the top ‘n’ results • Set a threshold for the entropy • Pick most likely senses • Ask: Did you mean … ?
Ranking with Sense • Simple filtering • More complex: • Term-sense frequency • IDF with respect to (term, sense)
Sense Diversification • Sense of top results vs. all relevant docs • If difference is large, suggest rare senses to user
Disambiguation Observations • WordNet senses are fine-grained • Small DA noise => large entropy noise • Short queries => ambiguity
Demo http://csil-linux40.cs.uiuc.edu:8080/