Mining Multiple Discriminative Patterns in Software Behavior Analysis

Hao Du, Chunping Li, Hui Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)


Sequence Classification has been a challenge task in recent years since sequence doesn’t have explicit features and the high-order temporal characteristics make the number of patterns extremely massive. Pattern-based classification has demonstrated its power in recent studies by mining discriminative features efficiently. Both binary and numerical discriminative features have been utilized for effective sequence classification, but the effect of each type of features hasn’t been analyzed separately. Our method selects the frequent closed unique iterative patterns as our candidate features, mined out the discriminative binary and numerical patterns for sequence classification, and given an insight into the discriminative power improvement by feature combinations. The experimental results on synthetic and real-life datasets reveal the validity of our approach.
Original languageEnglish
Title of host publicationInternational Conference on Ubiquitous Computing and Ambient Intelligence 2014: Proceedings
Place of PublicationSwitzerland
Number of pages8
Publication statusPublished - 2014
Externally publishedYes

Publication series

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

Bibliographical note

Ubiquitous Computing and Ambient Intelligence , UCamI 2014 ; Conference date: 02-12-2014 Through 05-12-2014


  • Data Mining
  • Multiple Discriminative Patterns
  • Software Behavior Analysis

Cite this