A generalisation of independence in statistical models for categorical distribution

    Research output: Contribution to journalArticle

    3 Citations (Scopus)

    Abstract

    In this paper, generalised statistical independence in statistical models for categorical distributions is proposed from the viewpoint of generalised multiplication characterised by a monotonically increasing function and its inverse function, and it is implemented in naive Bayes models. This paper also proposes an idea of their estimation method which directly uses empirical marginal distributions to retain simplicity of calculation. This method is interpreted as an optimisation of a rough approximation of the Bregman divergence so that it is expected to have a kind of robust property. Effectiveness of proposed models is shown by numerical experiments on some benchmark datasets.

    Original languageEnglish
    Pages (from-to)172-187
    Number of pages16
    JournalInternational Journal of Data Mining, Modelling and Management
    Volume4
    Issue number2
    DOIs
    Publication statusPublished - 2012

    Fingerprint

    Categorical
    Statistical Model
    Bregman Divergence
    Statistical Independence
    Inverse function
    Naive Bayes
    Empirical Distribution
    Increasing Functions
    Marginal Distribution
    Rough
    Simplicity
    Multiplication
    Numerical Experiment
    Benchmark
    Optimization
    Approximation
    Model
    Experiments
    Independence
    Generalization

    Keywords

    • Bregman divergence
    • Copula
    • Generalised independence
    • Independent model
    • Naive Bayes model

    ASJC Scopus subject areas

    • Computer Science Applications
    • Management Information Systems
    • Modelling and Simulation

    Cite this

    @article{9ad6b9a11794410780d6e592583a79e9,
    title = "A generalisation of independence in statistical models for categorical distribution",
    abstract = "In this paper, generalised statistical independence in statistical models for categorical distributions is proposed from the viewpoint of generalised multiplication characterised by a monotonically increasing function and its inverse function, and it is implemented in naive Bayes models. This paper also proposes an idea of their estimation method which directly uses empirical marginal distributions to retain simplicity of calculation. This method is interpreted as an optimisation of a rough approximation of the Bregman divergence so that it is expected to have a kind of robust property. Effectiveness of proposed models is shown by numerical experiments on some benchmark datasets.",
    keywords = "Bregman divergence, Copula, Generalised independence, Independent model, Naive Bayes model",
    author = "Yu Fujimoto and Noboru Murata",
    year = "2012",
    doi = "10.1504/IJDMMM.2012.046809",
    language = "English",
    volume = "4",
    pages = "172--187",
    journal = "International Journal of Data Mining, Modelling and Management",
    issn = "1759-1163",
    publisher = "Inderscience Publishers",
    number = "2",

    }

    TY - JOUR

    T1 - A generalisation of independence in statistical models for categorical distribution

    AU - Fujimoto, Yu

    AU - Murata, Noboru

    PY - 2012

    Y1 - 2012

    N2 - In this paper, generalised statistical independence in statistical models for categorical distributions is proposed from the viewpoint of generalised multiplication characterised by a monotonically increasing function and its inverse function, and it is implemented in naive Bayes models. This paper also proposes an idea of their estimation method which directly uses empirical marginal distributions to retain simplicity of calculation. This method is interpreted as an optimisation of a rough approximation of the Bregman divergence so that it is expected to have a kind of robust property. Effectiveness of proposed models is shown by numerical experiments on some benchmark datasets.

    AB - In this paper, generalised statistical independence in statistical models for categorical distributions is proposed from the viewpoint of generalised multiplication characterised by a monotonically increasing function and its inverse function, and it is implemented in naive Bayes models. This paper also proposes an idea of their estimation method which directly uses empirical marginal distributions to retain simplicity of calculation. This method is interpreted as an optimisation of a rough approximation of the Bregman divergence so that it is expected to have a kind of robust property. Effectiveness of proposed models is shown by numerical experiments on some benchmark datasets.

    KW - Bregman divergence

    KW - Copula

    KW - Generalised independence

    KW - Independent model

    KW - Naive Bayes model

    UR - http://www.scopus.com/inward/record.url?scp=84870288599&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84870288599&partnerID=8YFLogxK

    U2 - 10.1504/IJDMMM.2012.046809

    DO - 10.1504/IJDMMM.2012.046809

    M3 - Article

    AN - SCOPUS:84870288599

    VL - 4

    SP - 172

    EP - 187

    JO - International Journal of Data Mining, Modelling and Management

    JF - International Journal of Data Mining, Modelling and Management

    SN - 1759-1163

    IS - 2

    ER -