An algorithm for directed graph estimation

Hideitsu Hino, Atsushi Noda, Masami Tatsuno, Shotaro Akaho, Noboru Murata

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

    Abstract

    A problem of estimating the intrinsic graph structure from observed data is considered. The observed data in this study is a matrix with elements representing dependency between nodes in the graph. Each element of the observed matrix represents, for example, co-occurrence of events at two nodes, or correlation of variables corresponding to two nodes. The dependency does not represent direct connections and includes influences of various paths, and spurious correlations make the estimation of direct connection difficult. To alleviate this difficulty, digraph Laplacian is used for characterizing a graph. A generative model of an observed matrix is proposed, and a parameter estimation algorithm for the model is also proposed. The proposed method is capable of dealing with directed graphs, while conventional graph structure estimation methods from an observed matrix are only applicable to undirected graphs. Experimental result shows that the proposed algorithm is able to identify the intrinsic graph structure.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    PublisherSpringer Verlag
    Pages145-152
    Number of pages8
    Volume8681 LNCS
    ISBN (Print)9783319111780
    DOIs
    Publication statusPublished - 2014
    Event24th International Conference on Artificial Neural Networks, ICANN 2014 - Hamburg, Germany
    Duration: 2014 Sep 152014 Sep 19

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume8681 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other24th International Conference on Artificial Neural Networks, ICANN 2014
    CountryGermany
    CityHamburg
    Period14/9/1514/9/19

    Fingerprint

    Directed graphs
    Directed Graph
    Graph in graph theory
    Vertex of a graph
    Generative Models
    Parameter estimation
    Estimation Algorithms
    Undirected Graph
    Digraph
    Parameter Estimation
    Path
    Experimental Results

    Keywords

    • digraph Laplacian
    • directed graph
    • graph estimation

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Hino, H., Noda, A., Tatsuno, M., Akaho, S., & Murata, N. (2014). An algorithm for directed graph estimation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8681 LNCS, pp. 145-152). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8681 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-11179-7_19

    An algorithm for directed graph estimation. / Hino, Hideitsu; Noda, Atsushi; Tatsuno, Masami; Akaho, Shotaro; Murata, Noboru.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8681 LNCS Springer Verlag, 2014. p. 145-152 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8681 LNCS).

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

    Hino, H, Noda, A, Tatsuno, M, Akaho, S & Murata, N 2014, An algorithm for directed graph estimation. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8681 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8681 LNCS, Springer Verlag, pp. 145-152, 24th International Conference on Artificial Neural Networks, ICANN 2014, Hamburg, Germany, 14/9/15. https://doi.org/10.1007/978-3-319-11179-7_19
    Hino H, Noda A, Tatsuno M, Akaho S, Murata N. An algorithm for directed graph estimation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8681 LNCS. Springer Verlag. 2014. p. 145-152. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-11179-7_19
    Hino, Hideitsu ; Noda, Atsushi ; Tatsuno, Masami ; Akaho, Shotaro ; Murata, Noboru. / An algorithm for directed graph estimation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8681 LNCS Springer Verlag, 2014. pp. 145-152 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{9bb09960ea6d42b7964de91a37f7f6f0,
    title = "An algorithm for directed graph estimation",
    abstract = "A problem of estimating the intrinsic graph structure from observed data is considered. The observed data in this study is a matrix with elements representing dependency between nodes in the graph. Each element of the observed matrix represents, for example, co-occurrence of events at two nodes, or correlation of variables corresponding to two nodes. The dependency does not represent direct connections and includes influences of various paths, and spurious correlations make the estimation of direct connection difficult. To alleviate this difficulty, digraph Laplacian is used for characterizing a graph. A generative model of an observed matrix is proposed, and a parameter estimation algorithm for the model is also proposed. The proposed method is capable of dealing with directed graphs, while conventional graph structure estimation methods from an observed matrix are only applicable to undirected graphs. Experimental result shows that the proposed algorithm is able to identify the intrinsic graph structure.",
    keywords = "digraph Laplacian, directed graph, graph estimation",
    author = "Hideitsu Hino and Atsushi Noda and Masami Tatsuno and Shotaro Akaho and Noboru Murata",
    year = "2014",
    doi = "10.1007/978-3-319-11179-7_19",
    language = "English",
    isbn = "9783319111780",
    volume = "8681 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "145--152",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - An algorithm for directed graph estimation

    AU - Hino, Hideitsu

    AU - Noda, Atsushi

    AU - Tatsuno, Masami

    AU - Akaho, Shotaro

    AU - Murata, Noboru

    PY - 2014

    Y1 - 2014

    N2 - A problem of estimating the intrinsic graph structure from observed data is considered. The observed data in this study is a matrix with elements representing dependency between nodes in the graph. Each element of the observed matrix represents, for example, co-occurrence of events at two nodes, or correlation of variables corresponding to two nodes. The dependency does not represent direct connections and includes influences of various paths, and spurious correlations make the estimation of direct connection difficult. To alleviate this difficulty, digraph Laplacian is used for characterizing a graph. A generative model of an observed matrix is proposed, and a parameter estimation algorithm for the model is also proposed. The proposed method is capable of dealing with directed graphs, while conventional graph structure estimation methods from an observed matrix are only applicable to undirected graphs. Experimental result shows that the proposed algorithm is able to identify the intrinsic graph structure.

    AB - A problem of estimating the intrinsic graph structure from observed data is considered. The observed data in this study is a matrix with elements representing dependency between nodes in the graph. Each element of the observed matrix represents, for example, co-occurrence of events at two nodes, or correlation of variables corresponding to two nodes. The dependency does not represent direct connections and includes influences of various paths, and spurious correlations make the estimation of direct connection difficult. To alleviate this difficulty, digraph Laplacian is used for characterizing a graph. A generative model of an observed matrix is proposed, and a parameter estimation algorithm for the model is also proposed. The proposed method is capable of dealing with directed graphs, while conventional graph structure estimation methods from an observed matrix are only applicable to undirected graphs. Experimental result shows that the proposed algorithm is able to identify the intrinsic graph structure.

    KW - digraph Laplacian

    KW - directed graph

    KW - graph estimation

    UR - http://www.scopus.com/inward/record.url?scp=84958542694&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84958542694&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-11179-7_19

    DO - 10.1007/978-3-319-11179-7_19

    M3 - Conference contribution

    SN - 9783319111780

    VL - 8681 LNCS

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 145

    EP - 152

    BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    PB - Springer Verlag

    ER -