Minimizing the error of linear separators on linearly inseparable data

Boris Aronov, Delia Garijo, Yurai Nez-Rodrguez, David Rappaport, Carlos Seara, Jorge Urrutia

    Research output: Contribution to journalArticle

    Abstract

    Given linearly inseparable sets R of red points and B of blue points, we consider several measures of how far they are from being separable. Intuitively, given a potential separator ("classifier"), we measure its quality ("error") according to how much work it would take to move the misclassified points across the classifier to yield separated sets. We consider several measures of work and provide algorithms to find linear classifiers that minimize the error under these different measures.

    Original languageEnglish (US)
    Pages (from-to)1441-1452
    Number of pages12
    JournalDiscrete Applied Mathematics
    Volume160
    Issue number10-11
    DOIs
    StatePublished - Jul 2012

    Fingerprint

    Separator
    Separators
    Classifiers
    Linearly
    Classifier
    Quality Measures
    Minimise

    Keywords

    • Classifiers
    • Error minimizers
    • Linearly inseparable

    ASJC Scopus subject areas

    • Applied Mathematics
    • Discrete Mathematics and Combinatorics

    Cite this

    Aronov, B., Garijo, D., Nez-Rodrguez, Y., Rappaport, D., Seara, C., & Urrutia, J. (2012). Minimizing the error of linear separators on linearly inseparable data. Discrete Applied Mathematics, 160(10-11), 1441-1452. https://doi.org/10.1016/j.dam.2012.03.009

    Minimizing the error of linear separators on linearly inseparable data. / Aronov, Boris; Garijo, Delia; Nez-Rodrguez, Yurai; Rappaport, David; Seara, Carlos; Urrutia, Jorge.

    In: Discrete Applied Mathematics, Vol. 160, No. 10-11, 07.2012, p. 1441-1452.

    Research output: Contribution to journalArticle

    Aronov, B, Garijo, D, Nez-Rodrguez, Y, Rappaport, D, Seara, C & Urrutia, J 2012, 'Minimizing the error of linear separators on linearly inseparable data', Discrete Applied Mathematics, vol. 160, no. 10-11, pp. 1441-1452. https://doi.org/10.1016/j.dam.2012.03.009
    Aronov B, Garijo D, Nez-Rodrguez Y, Rappaport D, Seara C, Urrutia J. Minimizing the error of linear separators on linearly inseparable data. Discrete Applied Mathematics. 2012 Jul;160(10-11):1441-1452. https://doi.org/10.1016/j.dam.2012.03.009
    Aronov, Boris ; Garijo, Delia ; Nez-Rodrguez, Yurai ; Rappaport, David ; Seara, Carlos ; Urrutia, Jorge. / Minimizing the error of linear separators on linearly inseparable data. In: Discrete Applied Mathematics. 2012 ; Vol. 160, No. 10-11. pp. 1441-1452.
    @article{e7fe18bd336d459a819346594e2b40c2,
    title = "Minimizing the error of linear separators on linearly inseparable data",
    abstract = "Given linearly inseparable sets R of red points and B of blue points, we consider several measures of how far they are from being separable. Intuitively, given a potential separator ({"}classifier{"}), we measure its quality ({"}error{"}) according to how much work it would take to move the misclassified points across the classifier to yield separated sets. We consider several measures of work and provide algorithms to find linear classifiers that minimize the error under these different measures.",
    keywords = "Classifiers, Error minimizers, Linearly inseparable",
    author = "Boris Aronov and Delia Garijo and Yurai Nez-Rodrguez and David Rappaport and Carlos Seara and Jorge Urrutia",
    year = "2012",
    month = "7",
    doi = "10.1016/j.dam.2012.03.009",
    language = "English (US)",
    volume = "160",
    pages = "1441--1452",
    journal = "Discrete Applied Mathematics",
    issn = "0166-218X",
    publisher = "Elsevier",
    number = "10-11",

    }

    TY - JOUR

    T1 - Minimizing the error of linear separators on linearly inseparable data

    AU - Aronov, Boris

    AU - Garijo, Delia

    AU - Nez-Rodrguez, Yurai

    AU - Rappaport, David

    AU - Seara, Carlos

    AU - Urrutia, Jorge

    PY - 2012/7

    Y1 - 2012/7

    N2 - Given linearly inseparable sets R of red points and B of blue points, we consider several measures of how far they are from being separable. Intuitively, given a potential separator ("classifier"), we measure its quality ("error") according to how much work it would take to move the misclassified points across the classifier to yield separated sets. We consider several measures of work and provide algorithms to find linear classifiers that minimize the error under these different measures.

    AB - Given linearly inseparable sets R of red points and B of blue points, we consider several measures of how far they are from being separable. Intuitively, given a potential separator ("classifier"), we measure its quality ("error") according to how much work it would take to move the misclassified points across the classifier to yield separated sets. We consider several measures of work and provide algorithms to find linear classifiers that minimize the error under these different measures.

    KW - Classifiers

    KW - Error minimizers

    KW - Linearly inseparable

    UR - http://www.scopus.com/inward/record.url?scp=84859902279&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84859902279&partnerID=8YFLogxK

    U2 - 10.1016/j.dam.2012.03.009

    DO - 10.1016/j.dam.2012.03.009

    M3 - Article

    VL - 160

    SP - 1441

    EP - 1452

    JO - Discrete Applied Mathematics

    JF - Discrete Applied Mathematics

    SN - 0166-218X

    IS - 10-11

    ER -