Topic segmentation with an aspect hidden Markov model(3)

We present a novel probabilistic method for partially unsupervised topic segmentation on unstructured text. Previous approaches to this problem utilize the hidden Markov model framework (HMM). The HMM treats a document as mutually independent sets of words

Authorsemail:blei@cs.berkeley.edu,Pedro.Moreno@

cCompaqComputerCorporation,2001

Thisworkmaynotbecopiedorreproducedinwholeorinpartforanycommercialpurpose.Per-missiontocopyinwholeorinpartwithoutpaymentoffeeisgrantedfornonpro teducationalandresearchpurposesprovidedthatallsuchwholeorpartialcopiesincludethefollowing:anoticethatsuchcopyingisbypermissionoftheCambridgeResearchLaboratoryofCompaqComputerCorpo-rationinCambridge,Massachusetts;anacknowledgmentoftheauthorsandindividualcontributorstothework;andallapplicableportionsofthecopyrightnotice.Copying,reproducing,orrepub-lishingforanyotherpurposeshallrequirealicensewithpaymentoffeetotheCambridgeResearchLaboratory.Allrightsreserved.

CRLTechnicalreportsareavailableontheCRL’swebpageat

.

CompaqComputerCorporation

CambridgeResearchLaboratory

OneCambridgeCenter

Cambridge,Massachusetts02142USA