Graph-incorporated active learning with SVM

Jun Jiang, Horace H. S. Ip, Guilin Zhang

    Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

    Abstract

    Active learning is typically limited by the small sample problem which makes the resulting classifiers perform poorly, especially in the initial stages, To overcome this problem, in this paper, a novel framework - graph-incorporated active learning - is proposed, in which the selection pool is regarded as a graph. Its graph structure is applied to both improve sample selection criterion and provide the learner enough pseudo-labeled samples. By comparing with the state-of-the-art technique, the experiments on benchmark datasets show that the improvement of the proposed method is significant, i.e., it can solve the small problem well. The framework is combined with, but is not limited to, SVM.
    Original languageEnglish
    Title of host publicationMIPPR 2011
    Subtitle of host publicationRemote Sensing Image Processing, Geographic Information Systems, and Other Applications
    DOIs
    Publication statusPublished - Nov 2011
    Event7th International Symposium on Multispectral Image Processing & Pattern Recognition (MIPPR 2011) - Guilin, Hangzhou, China
    Duration: 4 Nov 20116 Nov 2011

    Publication series

    NameProceedings of SPIE - The International Society for Optical Engineering
    Volume8006
    ISSN (Print)0277-786X

    Conference

    Conference7th International Symposium on Multispectral Image Processing & Pattern Recognition (MIPPR 2011)
    PlaceChina
    CityGuilin, Hangzhou
    Period4/11/116/11/11

    Research Keywords

    • Active learning
    • Graph-based
    • Semi-supervised learning
    • Small sample problem
    • SVM

    Fingerprint

    Dive into the research topics of 'Graph-incorporated active learning with SVM'. Together they form a unique fingerprint.

    Cite this