Publication Details

VGEN: Fast Vertical Mining of Sequential Generator Patterns

FOURNIER-VIGER, P.; GOMARIZ, A.; ŠEBEK, M.; HLOSTA, M. VGEN: Fast Vertical Mining of Sequential Generator Patterns. In Data Warehousing and Knowledge Discovery. Munich: Springer Verlag, 2014. p. 476-488. ISBN: 978-3-319-10159-0.
Czech title
VGEN: Rychlé vertikální dolování sekvenčních generátorů
Type
conference paper
Language
English
Authors
Fournier-Viger Philippe (FIT)
Gomariz Antonio (FIT)
Šebek Michal, Ing., Ph.D.
Hlosta Martin, Ing., Ph.D.
URL
Keywords

sequential patterns, generators, vertical mining, candidate pruning

Abstract

Sequential pattern mining is a popular data mining task with wide applications. However, the set of all sequential patterns can be very large. To discover fewer but more representative patterns, several compact representations of sequential patterns have been studied. The set of sequential generatorsis one the most popular representations. It was shown to provide higher accuracy for classification than using all or only closed sequential patterns. Furthermore, mining generators is a key step in several other data mining tasks such as sequential rule generation. However, mining generators is computationally expensive. To address this issue, we propose a novel mining algorithm namedVGEN (Vertical sequential GENerator miner). An experimental study on five real datasets shows that VGEN is up to two orders of magnitude faster than the state-of-the-art algorithms for sequential generator mining.

Published
2014
Pages
476–488
Proceedings
Data Warehousing and Knowledge Discovery
ISBN
978-3-319-10159-0
Publisher
Springer Verlag
Place
Munich
DOI
EID Scopus
BibTeX
@inproceedings{BUT111554,
  author="Philippe {Fournier-Viger} and Antonio {Gomariz} and Michal {Šebek} and Martin {Hlosta}",
  title="VGEN: Fast Vertical Mining of Sequential Generator Patterns",
  booktitle="Data Warehousing and Knowledge Discovery",
  year="2014",
  pages="476--488",
  publisher="Springer Verlag",
  address="Munich",
  doi="10.1007/978-3-319-10160-6\{_}42",
  isbn="978-3-319-10159-0",
  url="http://dx.doi.org/10.1007/978-3-319-10160-6_42"
}
Back to top