Joint Structured Graph Learning and Unsupervised Feature Selection

Yong Peng, Leijie Zhang, Wanzeng Kong, Feiping Nie, Andrzej Cichocki

    Результат исследований: Глава в книге, отчете, сборнике статейМатериалы для конференциирецензирование

    1 Цитирования (Scopus)

    Аннотация

    The central task in graph-based unsupervised feature selection (GUFS) depends on two folds, one is to accurately characterize the geometrical structure of the original feature space with a graph and the other is to make the selected features well preserve such intrinsic structure. Currently, most of the existing GUFS methods use a two-stage strategy which constructs graph first and then perform feature selection on this fixed graph. Since the performance of feature selection severely depends on the quality of graph, the selection results will be unsatisfactory if the given graph is of low-quality. To this end, we propose a joint graph learning and unsupervised feature selection (JGUFS) model in which the graph can be adjusted to adapt the feature selection process. The JGUFS objective function is optimized by an efficient iterative algorithm whose convergence and complexity are analyzed in detail. Experimental results on representative benchmark data sets demonstrate the improved performance of JGUFS in comparison with state-of-the-art methods and therefore we conclude that it is promising of allowing the feature selection process to change the data graph.

    Язык оригиналаАнглийский
    Название основной публикации2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
    ИздательInstitute of Electrical and Electronics Engineers Inc.
    Страницы3572-3576
    Число страниц5
    ISBN (электронное издание)9781479981311
    DOI
    СостояниеОпубликовано - мая 2019
    Событие44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, Великобритания
    Продолжительность: 12 мая 201917 мая 2019

    Серия публикаций

    НазваниеICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
    Том2019-May
    ISSN (печатное издание)1520-6149

    Конференция

    Конференция44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
    Страна/TерриторияВеликобритания
    ГородBrighton
    Период12/05/1917/05/19

    Fingerprint

    Подробные сведения о темах исследования «Joint Structured Graph Learning and Unsupervised Feature Selection». Вместе они формируют уникальный семантический отпечаток (fingerprint).

    Цитировать