Dawn Lawrie
lawrie@cs.umass.edu
www-ciir.cs.umass.edu/~lawrie/

Title:
Finding Topic Terms for Hierarchical Summarization

Abstract:
Hierarchies have long been used for organization, summarization, and access
to information. In this talk I define summarization in terms of a
probabilistic language model and use the definition to explore a new
technique for automatically generating topic hierarchies.  This talk will
focus on the method for selecting terms, which interprets the language
model as a graph and applies an approximation to Dominating Set Problem.
This graph-theoretic problem captures the main ideas about what makes a
term a good topic term and allows us to efficiently generates topic terms
according to a language model.