Deakin University
Browse

File(s) under permanent embargo

Clustering based multi-label classification for image annotation and retrieval

conference contribution
posted on 2009-01-01, 00:00 authored by Gulisong NasierdingGulisong Nasierding, G Tsoumakas, Abbas KouzaniAbbas Kouzani
This paper presents a novel multi-label classification framework for domains with large numbers of labels. Automatic image annotation is such a domain, as the available semantic concepts are typically hundreds. The proposed framework comprises an initial clustering phase that breaks the original training set into several disjoint clusters of data. It then trains a multi-label classifier from the data of each cluster. Given a new test instance, the framework first finds the nearest cluster and then applies the corresponding model. Empirical results using two clustering algorithms, four multi-label classification algorithms and three image annotation data sets suggest that the proposed approach can improve the performance and reduce the training time of standard multi-label classification algorithms, particularly in the case of large number of labels.

History

Event

IEEE International Conference on Systems, Man, and Cybernetics (2009 : San Antonio, Texas)

Pagination

4514 - 4519

Publisher

IEEE

Location

San Antonio, Texas

Place of publication

Piscataway, N. J.

Start date

2009-10-11

End date

2009-10-14

ISSN

1062-922X

ISBN-13

9781424427932

Language

eng

Publication classification

E1 Full written paper - refereed

Copyright notice

2009, IEEE

Title of proceedings

SMC 2009 : Proceedings of the IEEE International Conference on Systems, Man and Cybernetics

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC