Determine the optimal parameter for information bottleneck method

Gang Li, Dong Liu, Yangdong Ye, Jia Rong

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review


A natural question in Information Bottleneck method is how many "groups" are appropriate. The dependency on prior knowledge restricts the applications of many Information Bottleneck algorithms. In this paper we aim to remove this dependency by formulating the parameter choosing as a model selection problem, and solve it using the minimum message length principle. Empirical results in the documentation clustering scenario indicates that the proposed method works well for the determination of the optimal parameter value for information bottleneck method.

Original languageEnglish
Title of host publicationPRICAI 2006
Subtitle of host publicationTrends in Artificial Intelligence - 9th Pacific Rim International Conference on Artificial Intelligence, Proceedings
PublisherSpringer-Verlag London Ltd.
Number of pages5
ISBN (Print)3540366679, 9783540366676
Publication statusPublished - 1 Jan 2006
EventPacific Rim International Conference on Artificial Intelligence 2006 - Guilin, China
Duration: 7 Aug 200611 Aug 2006
Conference number: 9th (Proceedings)

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferencePacific Rim International Conference on Artificial Intelligence 2006
Abbreviated titlePRICAI 2006
Internet address

Cite this