Abstract
A method for computationally visualizing and interpreting a text or corpus of texts in a taxonomy ofthe field is described. The method involves such stages as matching taxonomy topics and text(s) by using annotatedsuffix trees (ASTs), combining multiple information such as text abstracts, key-words and taxonomy crossreferences,building clusters of taxonomy topics and their profiles, and lifting the profiles to higher ranks of thetaxonomy hierarchy.
--------
URL address: http://ceur-ws.org/Vol-757/paper_3.pdf
--------
URL address: http://ceur-ws.org/Vol-757/paper_3.pdf
Original language | Unknown |
---|---|
Title of host publication | CEUR-WS |
Pages | 20-30 |
Publication status | Published - 1 Jan 2011 |
Event | Workshop on Concept Discovery in Unstructured Data (CDUD-2011) - Duration: 1 Jan 2011 → … |
Conference
Conference | Workshop on Concept Discovery in Unstructured Data (CDUD-2011) |
---|---|
Period | 1/01/11 → … |