Removing statistical biases in unsupervised sequence learning

Yoav Horman, Gal A. Kaminka

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

1 Scopus citations

Abstract

Unsupervised sequence learning is important to many applications. A learner is presented with unlabeled sequential data, and must discover sequential patterns that characterize the data. Popular approaches to such learning include statistical analysis and frequency based methods. We empirically compare these approaches and find that both approaches suffer from biases toward shorter sequences, and from inability to group together multiple instances of the same pattern. We provide methods to address these deficiencies, and evaluate them extensively on several synthetic and real-world data sets. The results show significant improvements in all learning methods used.

Original languageEnglish
Title of host publicationAdvances in Intelligent Data Analysis VI - 6th International Symposium on Intelligent Data Analysis, IDA 2005, Proceedings
PublisherSpringer Verlag
Pages157-167
Number of pages11
ISBN (Print)3540287957, 9783540287957
DOIs
StatePublished - 2005
Event6th International Symposium on Intelligent Data Analysis, IDA 2005 - Madrid, Spain
Duration: 8 Sep 200510 Sep 2005

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3646 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th International Symposium on Intelligent Data Analysis, IDA 2005
Country/TerritorySpain
CityMadrid
Period8/09/0510/09/05

Fingerprint

Dive into the research topics of 'Removing statistical biases in unsupervised sequence learning'. Together they form a unique fingerprint.

Cite this