Modelling Decision Making with Probabilistic Causation

DI Group Author

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Humans know how to reason based on cause and effect, but these are not enough to draw conclusions due to imperfect information and uncertainty. To address this, humans reason combining causal models with probabilistic information. The theory modeling causality and probability is known as Causal Bayes Nets - CBNs. We adopt a logic programming framework and methods to model our functional description of CBNs building on its many strengths and advantages. ACORDA is a declarative prospective logic programming system that simulates human reasoning in multiple steps into the future, but is not equipped to deal with probabilistic theory. P-log is a declarative logic programming language that can reason with probabilistic models. Integrated with P-log, ACORDA becomes ready to deal with uncertainty problems we face on a daily basis. We show how the integration of ACORDA and P-log has been accomplished, and present cases of daily life examples that ACORDA can help people to reason about.
Original languageUnknown
Pages (from-to)133-148
JournalIntelligent Decision Technologies
Volume4
Issue number2
DOIs
Publication statusPublished - 1 Jan 2010

Keywords

    Cite this

    DI Group Author. / Modelling Decision Making with Probabilistic Causation. In: Intelligent Decision Technologies. 2010 ; Vol. 4, No. 2. pp. 133-148.
    @article{719e5a4f8cc940ebb3b768b06ab925c8,
    title = "Modelling Decision Making with Probabilistic Causation",
    abstract = "Humans know how to reason based on cause and effect, but these are not enough to draw conclusions due to imperfect information and uncertainty. To address this, humans reason combining causal models with probabilistic information. The theory modeling causality and probability is known as Causal Bayes Nets - CBNs. We adopt a logic programming framework and methods to model our functional description of CBNs building on its many strengths and advantages. ACORDA is a declarative prospective logic programming system that simulates human reasoning in multiple steps into the future, but is not equipped to deal with probabilistic theory. P-log is a declarative logic programming language that can reason with probabilistic models. Integrated with P-log, ACORDA becomes ready to deal with uncertainty problems we face on a daily basis. We show how the integration of ACORDA and P-log has been accomplished, and present cases of daily life examples that ACORDA can help people to reason about.",
    keywords = "Bayes Nets, ACORDA, Prospective Logic Programming, Causal Models, P-log, Human Reasoning",
    author = "{DI Group Author}",
    note = "http://centria.di.fct.unl.pt/~lmp/publications/online-papers/IDT-Probabilistic-Causation.pdf",
    year = "2010",
    month = "1",
    day = "1",
    doi = "10.3233/IDT-2010-0074",
    language = "Unknown",
    volume = "4",
    pages = "133--148",
    journal = "Intelligent Decision Technologies",
    issn = "1872-4981",
    publisher = "IOS Press",
    number = "2",

    }

    Modelling Decision Making with Probabilistic Causation. / DI Group Author.

    In: Intelligent Decision Technologies, Vol. 4, No. 2, 01.01.2010, p. 133-148.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Modelling Decision Making with Probabilistic Causation

    AU - DI Group Author

    N1 - http://centria.di.fct.unl.pt/~lmp/publications/online-papers/IDT-Probabilistic-Causation.pdf

    PY - 2010/1/1

    Y1 - 2010/1/1

    N2 - Humans know how to reason based on cause and effect, but these are not enough to draw conclusions due to imperfect information and uncertainty. To address this, humans reason combining causal models with probabilistic information. The theory modeling causality and probability is known as Causal Bayes Nets - CBNs. We adopt a logic programming framework and methods to model our functional description of CBNs building on its many strengths and advantages. ACORDA is a declarative prospective logic programming system that simulates human reasoning in multiple steps into the future, but is not equipped to deal with probabilistic theory. P-log is a declarative logic programming language that can reason with probabilistic models. Integrated with P-log, ACORDA becomes ready to deal with uncertainty problems we face on a daily basis. We show how the integration of ACORDA and P-log has been accomplished, and present cases of daily life examples that ACORDA can help people to reason about.

    AB - Humans know how to reason based on cause and effect, but these are not enough to draw conclusions due to imperfect information and uncertainty. To address this, humans reason combining causal models with probabilistic information. The theory modeling causality and probability is known as Causal Bayes Nets - CBNs. We adopt a logic programming framework and methods to model our functional description of CBNs building on its many strengths and advantages. ACORDA is a declarative prospective logic programming system that simulates human reasoning in multiple steps into the future, but is not equipped to deal with probabilistic theory. P-log is a declarative logic programming language that can reason with probabilistic models. Integrated with P-log, ACORDA becomes ready to deal with uncertainty problems we face on a daily basis. We show how the integration of ACORDA and P-log has been accomplished, and present cases of daily life examples that ACORDA can help people to reason about.

    KW - Bayes Nets

    KW - ACORDA

    KW - Prospective Logic Programming

    KW - Causal Models

    KW - P-log

    KW - Human Reasoning

    U2 - 10.3233/IDT-2010-0074

    DO - 10.3233/IDT-2010-0074

    M3 - Article

    VL - 4

    SP - 133

    EP - 148

    JO - Intelligent Decision Technologies

    JF - Intelligent Decision Technologies

    SN - 1872-4981

    IS - 2

    ER -