Combining expert strategies in multimodal classification

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

Abstract

In multimodal classification, we look for a set of strategies for mining and exploiting the most informative modalities for a given situation. These strategies are computations performed by the algorithms. In this paper, we propose to consider strategies as advice given to an algorithm by "expert." There can be several classification strategies. Each strategy makes different assumptions regarding the fidelity of a sensor modality and uses different data to arrive at its estimates. Each strategy may place different trust in a sensor at different times, and each may be better in different situations. In this paper, we introduce a novel algorithm for combining expert strategies to achieve robust classification performance in a multimodal setting. We provide experimental results using real world examples to demonstrate the efficacy of the proposed algorithm.

Original languageEnglish
Title of host publicationInnovative Solutions in the Field of Engineering Sciences
PublisherTrans Tech Publications Ltd
Pages693-697
Number of pages5
ISBN (Print)9783038351511
DOIs
Publication statusPublished - 1 Jan 2014
Event2014 International Conference on Applied Mechanics and Mechanical Automation, AMMA 2014 - Macao, China
Duration: 20 May 201421 May 2014

Publication series

NameApplied Mechanics and Materials
Volume590
ISSN (Print)1660-9336
ISSN (Electronic)1662-7482

Other

Other2014 International Conference on Applied Mechanics and Mechanical Automation, AMMA 2014
CountryChina
CityMacao
Period20/05/1421/05/14

    Fingerprint

Keywords

  • Classification
  • Data fusion
  • Machine learning

Cite this

Peng, J. (2014). Combining expert strategies in multimodal classification. In Innovative Solutions in the Field of Engineering Sciences (pp. 693-697). (Applied Mechanics and Materials; Vol. 590). Trans Tech Publications Ltd. https://doi.org/10.4028/www.scientific.net/AMM.590.693