On active learning in hierarchical classification

Yu Cheng*, Kunpeng Zhang, Yusheng Xie, Ankit Agrawal, Alok Choudhary

*Corresponding author for this work

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

5 Scopus citations

Abstract

Most of the existing active learning algorithms assume all the category labels as independent or consider them in a "flat" structure. However, in reality, there are many applications in which the set of possible labels are often organized in a hierarchical structure. In this paper, we consider the problem of active learning when the categories are represented as a tree. Our goal is to exploit the structure information of the label tree in active learning to select the most informative samples to be labeled. We propose an algorithm that estimates the semantic space, embedding the category hierarchy. In this space, each category label is represented as a prototype and the uncertainty is measured using a variance-based fashion. We also demonstrate notable performance improvement with the proposed approach on synthetic and real datasets.

Original languageEnglish (US)
Title of host publicationCIKM 2012 - Proceedings of the 21st ACM International Conference on Information and Knowledge Management
Pages2468-2471
Number of pages4
DOIs
StatePublished - Dec 19 2012
Event21st ACM International Conference on Information and Knowledge Management, CIKM 2012 - Maui, HI, United States
Duration: Oct 29 2012Nov 2 2012

Publication series

NameACM International Conference Proceeding Series

Other

Other21st ACM International Conference on Information and Knowledge Management, CIKM 2012
CountryUnited States
CityMaui, HI
Period10/29/1211/2/12

Keywords

  • active learning
  • hierarchical classification
  • label tree embedding

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'On active learning in hierarchical classification'. Together they form a unique fingerprint.

Cite this