Leveraging node attributes for incomplete relational data

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    Abstract

    Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.
    Original languageEnglish
    Title of host publicationInternational Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia
    EditorsDoina Precup, Yee Whye Teh
    Place of PublicationUSA
    PublisherProceedings of Machine Learning Research (PMLR)
    Pages4072-4081
    Number of pages10
    Publication statusPublished - 2017
    EventInternational Conference on Machine Learning 2017 - International Convention Centre , Sydney , Australia
    Duration: 6 Aug 201711 Aug 2018
    Conference number: 34th
    https://icml.cc/Conferences/2017
    https://2017.icml.cc/

    Publication series

    NameProceedings of Machine Learning Research
    PublisherPMLR
    Volume70
    ISSN (Print)1938-7228

    Conference

    ConferenceInternational Conference on Machine Learning 2017
    Abbreviated titleICML 2017
    CountryAustralia
    CitySydney
    Period6/08/1711/08/18
    Internet address

    Cite this

    Zhao, H., Du, L., & Buntine, W. (2017). Leveraging node attributes for incomplete relational data. In D. Precup, & Y. W. Teh (Eds.), International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia (pp. 4072-4081). (Proceedings of Machine Learning Research; Vol. 70). USA: Proceedings of Machine Learning Research (PMLR).
    Zhao, He ; Du, Lan ; Buntine, Wray. / Leveraging node attributes for incomplete relational data. International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia. editor / Doina Precup ; Yee Whye Teh. USA : Proceedings of Machine Learning Research (PMLR), 2017. pp. 4072-4081 (Proceedings of Machine Learning Research).
    @inproceedings{e530c2e423ef424b814bec621b0d9c95,
    title = "Leveraging node attributes for incomplete relational data",
    abstract = "Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.",
    author = "He Zhao and Lan Du and Wray Buntine",
    year = "2017",
    language = "English",
    series = "Proceedings of Machine Learning Research",
    publisher = "Proceedings of Machine Learning Research (PMLR)",
    pages = "4072--4081",
    editor = "Doina Precup and Teh, {Yee Whye}",
    booktitle = "International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia",

    }

    Zhao, H, Du, L & Buntine, W 2017, Leveraging node attributes for incomplete relational data. in D Precup & YW Teh (eds), International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia. Proceedings of Machine Learning Research, vol. 70, Proceedings of Machine Learning Research (PMLR), USA, pp. 4072-4081, International Conference on Machine Learning 2017, Sydney , Australia, 6/08/17.

    Leveraging node attributes for incomplete relational data. / Zhao, He; Du, Lan; Buntine, Wray.

    International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia. ed. / Doina Precup; Yee Whye Teh. USA : Proceedings of Machine Learning Research (PMLR), 2017. p. 4072-4081 (Proceedings of Machine Learning Research; Vol. 70).

    Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

    TY - GEN

    T1 - Leveraging node attributes for incomplete relational data

    AU - Zhao, He

    AU - Du, Lan

    AU - Buntine, Wray

    PY - 2017

    Y1 - 2017

    N2 - Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.

    AB - Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood. Our method works flexibly with both directed and undirected relational networks. The inference can be done by efficient Gibbs sampling which leverages sparsity of both networks and node attributes. Extensive experiments show that our models achieve the state-of-the-art link prediction results, especially with highly incomplete relational data.

    M3 - Conference Paper

    T3 - Proceedings of Machine Learning Research

    SP - 4072

    EP - 4081

    BT - International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia

    A2 - Precup, Doina

    A2 - Teh, Yee Whye

    PB - Proceedings of Machine Learning Research (PMLR)

    CY - USA

    ER -

    Zhao H, Du L, Buntine W. Leveraging node attributes for incomplete relational data. In Precup D, Teh YW, editors, International Conference on Machine Learning, 6-11 August 2017, International Convention Centre, Sydney, Australia. USA: Proceedings of Machine Learning Research (PMLR). 2017. p. 4072-4081. (Proceedings of Machine Learning Research).