Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach

Paloma Lucena-Moya, Renee Brawata, Jarrod KATH, Evan HARRISON, Sondoss ElSawah, Fiona DYER

    Research output: Contribution to journalArticle

    10 Citations (Scopus)

    Abstract

    Bayesian networks (BNs) are a popular tool in natural resource management but are limited when dealing with ecological assemblage data and when discretizing continuous variables. We present a method that addresses these challenges using a BN model developed for the Upper Murrumbidgee River Catchment (south-eastern Australia). A selection process was conducted to choose the taxa from the whole macroinvertebrate assemblage that were incorporated in the BN as endpoints. Furthermore, two different approaches to the discretization of continuous predictor variables for the BN were compared. One approach used Threshold Indicator Taxa Analysis (TITAN) which estimates the thresholds based on the biological community. The other approach used was the expert opinion. The TITAN-based discretizations provided comparable predictions to expert opinion-based discretizations but in combining statistical rigor and ecological relevance, offer a novel and objective approach to the discretization. The TITAN-based method may be used together with expert opinion.
    Original languageEnglish
    Pages (from-to)36-45
    Number of pages10
    JournalEnvironmental Modelling and Software
    Volume66
    Issue number6
    DOIs
    Publication statusPublished - 2015

    Fingerprint

    Bayesian networks
    Natural resources management
    Catchments
    macroinvertebrate
    resource management
    natural resource
    Rivers
    catchment
    prediction
    river
    analysis
    opinion
    indicator
    method

    Cite this

    Lucena-Moya, Paloma ; Brawata, Renee ; KATH, Jarrod ; HARRISON, Evan ; ElSawah, Sondoss ; DYER, Fiona. / Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach. In: Environmental Modelling and Software. 2015 ; Vol. 66, No. 6. pp. 36-45.
    @article{0bdf611f11104c468a0b3a22355a162b,
    title = "Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach",
    abstract = "Bayesian networks (BNs) are a popular tool in natural resource management but are limited when dealing with ecological assemblage data and when discretizing continuous variables. We present a method that addresses these challenges using a BN model developed for the Upper Murrumbidgee River Catchment (south-eastern Australia). A selection process was conducted to choose the taxa from the whole macroinvertebrate assemblage that were incorporated in the BN as endpoints. Furthermore, two different approaches to the discretization of continuous predictor variables for the BN were compared. One approach used Threshold Indicator Taxa Analysis (TITAN) which estimates the thresholds based on the biological community. The other approach used was the expert opinion. The TITAN-based discretizations provided comparable predictions to expert opinion-based discretizations but in combining statistical rigor and ecological relevance, offer a novel and objective approach to the discretization. The TITAN-based method may be used together with expert opinion.",
    author = "Paloma Lucena-Moya and Renee Brawata and Jarrod KATH and Evan HARRISON and Sondoss ElSawah and Fiona DYER",
    year = "2015",
    doi = "10.1016/j.envsoft.2014.12.019",
    language = "English",
    volume = "66",
    pages = "36--45",
    journal = "Environmental Software",
    issn = "1364-8152",
    publisher = "Elsevier BV",
    number = "6",

    }

    Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach. / Lucena-Moya, Paloma; Brawata, Renee; KATH, Jarrod; HARRISON, Evan; ElSawah, Sondoss; DYER, Fiona.

    In: Environmental Modelling and Software, Vol. 66, No. 6, 2015, p. 36-45.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Discretization of continuous predictor variables in Bayesian networks: An ecological threshold approach

    AU - Lucena-Moya, Paloma

    AU - Brawata, Renee

    AU - KATH, Jarrod

    AU - HARRISON, Evan

    AU - ElSawah, Sondoss

    AU - DYER, Fiona

    PY - 2015

    Y1 - 2015

    N2 - Bayesian networks (BNs) are a popular tool in natural resource management but are limited when dealing with ecological assemblage data and when discretizing continuous variables. We present a method that addresses these challenges using a BN model developed for the Upper Murrumbidgee River Catchment (south-eastern Australia). A selection process was conducted to choose the taxa from the whole macroinvertebrate assemblage that were incorporated in the BN as endpoints. Furthermore, two different approaches to the discretization of continuous predictor variables for the BN were compared. One approach used Threshold Indicator Taxa Analysis (TITAN) which estimates the thresholds based on the biological community. The other approach used was the expert opinion. The TITAN-based discretizations provided comparable predictions to expert opinion-based discretizations but in combining statistical rigor and ecological relevance, offer a novel and objective approach to the discretization. The TITAN-based method may be used together with expert opinion.

    AB - Bayesian networks (BNs) are a popular tool in natural resource management but are limited when dealing with ecological assemblage data and when discretizing continuous variables. We present a method that addresses these challenges using a BN model developed for the Upper Murrumbidgee River Catchment (south-eastern Australia). A selection process was conducted to choose the taxa from the whole macroinvertebrate assemblage that were incorporated in the BN as endpoints. Furthermore, two different approaches to the discretization of continuous predictor variables for the BN were compared. One approach used Threshold Indicator Taxa Analysis (TITAN) which estimates the thresholds based on the biological community. The other approach used was the expert opinion. The TITAN-based discretizations provided comparable predictions to expert opinion-based discretizations but in combining statistical rigor and ecological relevance, offer a novel and objective approach to the discretization. The TITAN-based method may be used together with expert opinion.

    U2 - 10.1016/j.envsoft.2014.12.019

    DO - 10.1016/j.envsoft.2014.12.019

    M3 - Article

    VL - 66

    SP - 36

    EP - 45

    JO - Environmental Software

    JF - Environmental Software

    SN - 1364-8152

    IS - 6

    ER -