Abstract
This paper investigates the coding of change-points in the information-theoretic Minimum Message Length (MML) framework. Change-point coding regions affect model selection and parameter estimation in problems such as time series segmentation and decision trees. The Minimum Message Length (MML) and Minimum Description Length (MDL78) approaches to change-point problems have been shown to perform well by several authors. In this paper we compare some published MML and MDL78 methods and introduce some new MML approximations called ‘MMLDc’ and ‘MMLDF’. These new approximations are empirically compared with Strict MML (SMML), Fairly Strict MML (FSMML), MML68, the Minimum Expected Kullback-Leibler Distance (MEKLD) loss function and MDL78 on a tractable binomial change-point problem.
Original language | English |
---|---|
Title of host publication | PRICAI 2002: Trends in Artificial Intelligence |
Subtitle of host publication | 7th Pacific Rim International Conference on Artificial Intelligence Tokyo, Japan, August 18-22, 2002 Proceedings |
Editors | Mitsuru Ishizuka, Abdul Sattar |
Place of Publication | Berlin Germany |
Publisher | Springer |
Pages | 244-254 |
Number of pages | 11 |
ISBN (Print) | 3540440380 |
DOIs | |
Publication status | Published - 2002 |
Event | Pacific Rim International Conference on Artificial Intelligence 2002 - Tokyo, Japan Duration: 18 Aug 2002 → 22 Aug 2002 Conference number: 7th https://link.springer.com/book/10.1007/3-540-45683-X (Proceedings) |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Publisher | Springer |
Volume | 2417 |
ISSN (Print) | 0302-9743 |
Conference
Conference | Pacific Rim International Conference on Artificial Intelligence 2002 |
---|---|
Abbreviated title | PRICAI 2002 |
Country/Territory | Japan |
City | Tokyo |
Period | 18/08/02 → 22/08/02 |
Internet address |
|