Skip to main navigation Skip to search Skip to main content

MISUP: Multiple-instance learning via immunological suppression mechanism

  • Duzhou Zhang
  • , Xibin Cao*
  • *Corresponding author for this work
  • School of Astronautics, Harbin Institute of Technology
  • China Aerospace Science and Technology Corporation

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

Abstract

In multiple-instance learning (MIL), examples are sets of instances named bags and labels are associated with bags rather than instances. A bag is labeled as positive if it contains at least one positive instance; otherwise, labeled as negative. Recently, several instance selection-based MIL (ISMIL) algorithms show their power in solving the MIL problem. In this paper, we propose a new ISMIL algorithm based on the self-regulation and suppression mechanisms found in the biological immune system. Experimental results show that our MIL algorithm is highly comparable with other ISMIL ones in terms of classification accuracy and computation time.

Original languageEnglish
Title of host publicationNeural Information Processing - 20th International Conference, ICONIP 2013, Proceedings
Pages27-34
Number of pages8
EditionPART 2
DOIs
StatePublished - 2013
Externally publishedYes
Event20th International Conference on Neural Information Processing, ICONIP 2013 - Daegu, Korea, Republic of
Duration: 3 Nov 20137 Nov 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume8227 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference20th International Conference on Neural Information Processing, ICONIP 2013
Country/TerritoryKorea, Republic of
CityDaegu
Period3/11/137/11/13

Keywords

  • Artificial immune systems
  • Instance selection
  • Multiple-instance learning
  • Support vector machines

Fingerprint

Dive into the research topics of 'MISUP: Multiple-instance learning via immunological suppression mechanism'. Together they form a unique fingerprint.

Cite this