Markov blanket discovery in positive-unlabelled and semi-supervised data

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    The importance of Markov blanket discovery algorithms istwofold: as the main building block in constraint-based structure learn-ing of Bayesian network algorithms and as a technique to derive theoptimal set of features in filter feature selection approaches. Equally,learning from partially labelled data is a crucial and demanding area ofmachine learning, and extending techniques from fully to partially super-vised scenarios is a challenging problem. While there are many differentalgorithms to derive the Markov blanket of fully supervised nodes, thepartially-labelled problem is far more challenging, and there is a lack ofprincipled approaches in the literature. Our work derives a generaliza-tion of the conditional tests of independence for partially labelled binarytarget variables, which can handle the two main partially labelled scenar-ios:positive-unlabelled and semi-supervised.The result is a significantlydeeper understanding of how to control false negative errors in MarkovBlanket discovery procedures and how unlabelled data can help.
    Original languageEnglish
    Title of host publicationProceedings of European Conference, ECML PKDD 2015 Part 1
    PublisherSpringer Nature
    Pages351-366
    Number of pages16
    Volume9284
    ISBN (Print)978-3-319-23527-1
    DOIs
    Publication statusPublished - Sep 2015
    EventEuropean Conference on Machine Learning - ECML/PKDD 2015 - Porto, Portugal
    Duration: 7 Sep 201511 Sep 2015

    Publication series

    NameLecture Notes in Computer Science

    Conference

    ConferenceEuropean Conference on Machine Learning - ECML/PKDD 2015
    CityPorto, Portugal
    Period7/09/1511/09/15

    Keywords

    • Markov blanket discovery; Partially labelled; Positive un- labelled; Semi supervised; Mutual information

    Fingerprint

    Dive into the research topics of 'Markov blanket discovery in positive-unlabelled and semi-supervised data'. Together they form a unique fingerprint.

    Cite this