Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches

Morten Mørup, Fumiko Kano Glückstad, Tue Herlau, Mikkel Nørgaard Schmidt

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    Abstract

    Structuring knowledge systems with binary features is often based on imposing a similarity measure and clustering objects according to this similarity. Unfortunately, such analyses can be heavily influenced by the choice of similarity measure. Furthermore, it is unclear at which level clusters have statistical support and how this approach generalizes to the structuring and alignment of knowledge systems. We propose a non-parametric Bayesian generative model for structuring binary feature data that does not depend on a specific choice of similarity measure. We jointly model all combinations of binary matches and structure the data into groups at the level in which they have statistical support. The model naturally extends to structuring and aligning an arbitrary number of systems. We analyze three datasets on educational concepts and their features and demonstrate how the proposed model can both be used to structure each system separately or to jointly align two or more systems. The proposed method forms a promising new framework for the statistical modeling and alignment of structure across an arbitrary number of systems.
    Structuring knowledge systems with binary features is often based on imposing a similarity measure and clustering objects according to this similarity. Unfortunately, such analyses can be heavily influenced by the choice of similarity measure. Furthermore, it is unclear at which level clusters have statistical support and how this approach generalizes to the structuring and alignment of knowledge systems. We propose a non-parametric Bayesian generative model for structuring binary feature data that does not depend on a specific choice of similarity measure. We jointly model all combinations of binary matches and structure the data into groups at the level in which they have statistical support. The model naturally extends to structuring and aligning an arbitrary number of systems. We analyze three datasets on educational concepts and their features and demonstrate how the proposed model can both be used to structure each system separately or to jointly align two or more systems. The proposed method forms a promising new framework for the statistical modeling and alignment of structure across an arbitrary number of systems.
    LanguageEnglish
    Title of host publicationProceedings of MLSP 2014
    EditorsMamadou Mboup, Tülay Adali, Éric Moreau, Jan Larsen
    Number of pages6
    Place of PublicationNew York
    PublisherIEEE
    Date2014
    ISBN (Print)9781479936946
    DOIs
    StatePublished - 2014
    EventThe 24th IEEE International Workshop on Machine Learning for Signal Processing. MLSP 2014 - Reims Centre des Congrès, Reims, France
    Duration: 21 Sep 201424 Sep 2014
    Conference number: 24
    http://mlsp2014.conwiz.dk/home.htm

    Conference

    ConferenceThe 24th IEEE International Workshop on Machine Learning for Signal Processing. MLSP 2014
    Number24
    LocationReims Centre des Congrès
    CountryFrance
    CityReims
    Period21/09/201424/09/2014
    Internet address

    Bibliographical note

    CBS Library does not have access to the material

    Keywords

    • Bayesian non-parametrics
    • Relational modeling
    • Binary similarity
    • Knowledge structuring

    Cite this

    Mørup, M., Kano Glückstad, F., Herlau, T., & Schmidt, M. N. (2014). Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches. In M. Mboup, T. Adali, É. Moreau, & J. Larsen (Eds.), Proceedings of MLSP 2014 New York: IEEE. DOI: 10.1109/MLSP.2014.6958905
    Mørup, Morten ; Kano Glückstad, Fumiko ; Herlau, Tue ; Schmidt, Mikkel Nørgaard. / Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches. Proceedings of MLSP 2014. editor / Mamadou Mboup ; Tülay Adali ; Éric Moreau ; Jan Larsen. New York : IEEE, 2014.
    @inproceedings{8331fec88f8e4823b49328f895d938e3,
    title = "Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches",
    abstract = "Structuring knowledge systems with binary features is often based on imposing a similarity measure and clustering objects according to this similarity. Unfortunately, such analyses can be heavily influenced by the choice of similarity measure. Furthermore, it is unclear at which level clusters have statistical support and how this approach generalizes to the structuring and alignment of knowledge systems. We propose a non-parametric Bayesian generative model for structuring binary feature data that does not depend on a specific choice of similarity measure. We jointly model all combinations of binary matches and structure the data into groups at the level in which they have statistical support. The model naturally extends to structuring and aligning an arbitrary number of systems. We analyze three datasets on educational concepts and their features and demonstrate how the proposed model can both be used to structure each system separately or to jointly align two or more systems. The proposed method forms a promising new framework for the statistical modeling and alignment of structure across an arbitrary number of systems.",
    keywords = "Bayesian non-parametrics, Relational modeling, Binary similarity, Knowledge structuring",
    author = "Morten M{\o}rup and {Kano Gl{\"u}ckstad}, Fumiko and Tue Herlau and Schmidt, {Mikkel N{\o}rgaard}",
    note = "CBS Library does not have access to the material",
    year = "2014",
    doi = "10.1109/MLSP.2014.6958905",
    language = "English",
    isbn = "9781479936946",
    editor = "Mamadou Mboup and T{\"u}lay Adali and {\'E}ric Moreau and Jan Larsen",
    booktitle = "Proceedings of MLSP 2014",
    publisher = "IEEE",
    address = "United States",

    }

    Mørup, M, Kano Glückstad, F, Herlau, T & Schmidt, MN 2014, Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches. in M Mboup, T Adali, É Moreau & J Larsen (eds), Proceedings of MLSP 2014. IEEE, New York, The 24th IEEE International Workshop on Machine Learning for Signal Processing. MLSP 2014, Reims, France, 21/09/2014. DOI: 10.1109/MLSP.2014.6958905

    Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches. / Mørup, Morten; Kano Glückstad, Fumiko ; Herlau, Tue; Schmidt, Mikkel Nørgaard.

    Proceedings of MLSP 2014. ed. / Mamadou Mboup; Tülay Adali; Éric Moreau; Jan Larsen. New York : IEEE, 2014.

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    TY - GEN

    T1 - Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches

    AU - Mørup,Morten

    AU - Kano Glückstad,Fumiko

    AU - Herlau,Tue

    AU - Schmidt,Mikkel Nørgaard

    N1 - CBS Library does not have access to the material

    PY - 2014

    Y1 - 2014

    N2 - Structuring knowledge systems with binary features is often based on imposing a similarity measure and clustering objects according to this similarity. Unfortunately, such analyses can be heavily influenced by the choice of similarity measure. Furthermore, it is unclear at which level clusters have statistical support and how this approach generalizes to the structuring and alignment of knowledge systems. We propose a non-parametric Bayesian generative model for structuring binary feature data that does not depend on a specific choice of similarity measure. We jointly model all combinations of binary matches and structure the data into groups at the level in which they have statistical support. The model naturally extends to structuring and aligning an arbitrary number of systems. We analyze three datasets on educational concepts and their features and demonstrate how the proposed model can both be used to structure each system separately or to jointly align two or more systems. The proposed method forms a promising new framework for the statistical modeling and alignment of structure across an arbitrary number of systems.

    AB - Structuring knowledge systems with binary features is often based on imposing a similarity measure and clustering objects according to this similarity. Unfortunately, such analyses can be heavily influenced by the choice of similarity measure. Furthermore, it is unclear at which level clusters have statistical support and how this approach generalizes to the structuring and alignment of knowledge systems. We propose a non-parametric Bayesian generative model for structuring binary feature data that does not depend on a specific choice of similarity measure. We jointly model all combinations of binary matches and structure the data into groups at the level in which they have statistical support. The model naturally extends to structuring and aligning an arbitrary number of systems. We analyze three datasets on educational concepts and their features and demonstrate how the proposed model can both be used to structure each system separately or to jointly align two or more systems. The proposed method forms a promising new framework for the statistical modeling and alignment of structure across an arbitrary number of systems.

    KW - Bayesian non-parametrics

    KW - Relational modeling

    KW - Binary similarity

    KW - Knowledge structuring

    U2 - 10.1109/MLSP.2014.6958905

    DO - 10.1109/MLSP.2014.6958905

    M3 - Article in proceedings

    SN - 9781479936946

    BT - Proceedings of MLSP 2014

    PB - IEEE

    CY - New York

    ER -

    Mørup M, Kano Glückstad F, Herlau T, Schmidt MN. Nonparametric Statistical Structuring of Knowledge Systems Using Binary Feature Matches. In Mboup M, Adali T, Moreau É, Larsen J, editors, Proceedings of MLSP 2014. New York: IEEE. 2014. Available from, DOI: 10.1109/MLSP.2014.6958905