Fine-Grained Hidden Markov Modeling for Broadcast-News Story Segmentation
March 2001
Warren Greiff, The MITRE Corporation
Alex Morgan, The MITRE Corporation
Randall Fish, The MITRE Corporation
Marc Richards, The MITRE Corporation
Amlan Kundu, The MITRE Corporation
ABSTRACT
We present the design and development of a Hidden Markov Model for the division of news broadcasts into story segments. Model topology, and the textual features used, are discussed, together with the non-parametric estimation techniques that were employed for obtaining estimates for both transition and observation probabilities. Visualization methods developed for the analysis of system performance are also presented.
Additional Search Keywords
n/a
|
