Patchworking multiple pairwise distances for learning with distance matrices

Ken Takano, Hideitsu Hino, Yuki Yoshikawa, Noboru Murata

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

    1 Citation (Scopus)

    Abstract

    A classification framework using only a set of distance matrices is proposed. The proposed algorithm can learn a classifier only from a set of distance matrices or similarity matrices, hence applicable to structured data, which do not have natural vector representation such as time series and graphs. Random forest is used to explore ideal feature representation based on the distance between points defined by a set of given distance matrices. The effectiveness of the proposed method is evaluated through experiments with point process data and graph structured data.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    PublisherSpringer Verlag
    Pages287-294
    Number of pages8
    Volume9237
    ISBN (Print)9783319224817
    DOIs
    Publication statusPublished - 2015
    Event12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015 - Liberec, Czech Republic
    Duration: 2015 Aug 252015 Aug 28

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume9237
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015
    CountryCzech Republic
    CityLiberec
    Period15/8/2515/8/28

    Fingerprint

    Distance Matrix
    Pairwise
    Random Forest
    Point Process
    Graph in graph theory
    Time series
    Classifier
    Classifiers
    Experiment
    Learning
    Experiments

    Keywords

    • Classification
    • Decision trees
    • Graph kernel
    • Random forest
    • Spike train
    • Structured data

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Takano, K., Hino, H., Yoshikawa, Y., & Murata, N. (2015). Patchworking multiple pairwise distances for learning with distance matrices. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9237, pp. 287-294). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9237). Springer Verlag. https://doi.org/10.1007/978-3-319-22482-4_33

    Patchworking multiple pairwise distances for learning with distance matrices. / Takano, Ken; Hino, Hideitsu; Yoshikawa, Yuki; Murata, Noboru.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9237 Springer Verlag, 2015. p. 287-294 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9237).

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

    Takano, K, Hino, H, Yoshikawa, Y & Murata, N 2015, Patchworking multiple pairwise distances for learning with distance matrices. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 9237, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9237, Springer Verlag, pp. 287-294, 12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015, Liberec, Czech Republic, 15/8/25. https://doi.org/10.1007/978-3-319-22482-4_33
    Takano K, Hino H, Yoshikawa Y, Murata N. Patchworking multiple pairwise distances for learning with distance matrices. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9237. Springer Verlag. 2015. p. 287-294. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-22482-4_33
    Takano, Ken ; Hino, Hideitsu ; Yoshikawa, Yuki ; Murata, Noboru. / Patchworking multiple pairwise distances for learning with distance matrices. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9237 Springer Verlag, 2015. pp. 287-294 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{ea99a91783a84408be8262060162f07c,
    title = "Patchworking multiple pairwise distances for learning with distance matrices",
    abstract = "A classification framework using only a set of distance matrices is proposed. The proposed algorithm can learn a classifier only from a set of distance matrices or similarity matrices, hence applicable to structured data, which do not have natural vector representation such as time series and graphs. Random forest is used to explore ideal feature representation based on the distance between points defined by a set of given distance matrices. The effectiveness of the proposed method is evaluated through experiments with point process data and graph structured data.",
    keywords = "Classification, Decision trees, Graph kernel, Random forest, Spike train, Structured data",
    author = "Ken Takano and Hideitsu Hino and Yuki Yoshikawa and Noboru Murata",
    year = "2015",
    doi = "10.1007/978-3-319-22482-4_33",
    language = "English",
    isbn = "9783319224817",
    volume = "9237",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "287--294",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - Patchworking multiple pairwise distances for learning with distance matrices

    AU - Takano, Ken

    AU - Hino, Hideitsu

    AU - Yoshikawa, Yuki

    AU - Murata, Noboru

    PY - 2015

    Y1 - 2015

    N2 - A classification framework using only a set of distance matrices is proposed. The proposed algorithm can learn a classifier only from a set of distance matrices or similarity matrices, hence applicable to structured data, which do not have natural vector representation such as time series and graphs. Random forest is used to explore ideal feature representation based on the distance between points defined by a set of given distance matrices. The effectiveness of the proposed method is evaluated through experiments with point process data and graph structured data.

    AB - A classification framework using only a set of distance matrices is proposed. The proposed algorithm can learn a classifier only from a set of distance matrices or similarity matrices, hence applicable to structured data, which do not have natural vector representation such as time series and graphs. Random forest is used to explore ideal feature representation based on the distance between points defined by a set of given distance matrices. The effectiveness of the proposed method is evaluated through experiments with point process data and graph structured data.

    KW - Classification

    KW - Decision trees

    KW - Graph kernel

    KW - Random forest

    KW - Spike train

    KW - Structured data

    UR - http://www.scopus.com/inward/record.url?scp=84944681542&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84944681542&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-22482-4_33

    DO - 10.1007/978-3-319-22482-4_33

    M3 - Conference contribution

    AN - SCOPUS:84944681542

    SN - 9783319224817

    VL - 9237

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

    SP - 287

    EP - 294

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

    PB - Springer Verlag

    ER -