Learning feature weights from positive cases

Sidath Gunawardena, Rosina O. Weber, Julia Stoyanovich

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse experiences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to recommend the characteristics of two collaborators to complement an academic seeking a multidisciplinary team; the three form a collaboration that resembles a configuration that has been successful in securing funding. While seeking a suitable measure for computing similarity between cases, we were confronted with two challenges: a problem context with insufficient domain knowledge and data that consists exclusively of successful collaborations, that is, it contains only positive instances. We present our strategy to overcome these challenges, which is a clustering-based approach to learn feature weights. Our approach identifies poorly aligned cases, i.e., ones that violate the assumption that similar problems have similar solutions. We use the poorly aligned cases as negatives in a feedback algorithm to learn feature weights. The result of this work is an integration of methods that makes CBR useful to yet another context and in conditions it has not been used before.

    Original languageEnglish (US)
    Title of host publicationCase-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings
    Pages134-148
    Number of pages15
    DOIs
    StatePublished - Sep 27 2013
    Event21st International Conference on Case-Based Reasoning Research and Development, ICCBR 2013 - Saratoga Springs, NY, United States
    Duration: Jul 8 2013Jul 11 2013

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume7969 LNAI
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Other

    Other21st International Conference on Case-Based Reasoning Research and Development, ICCBR 2013
    CountryUnited States
    CitySaratoga Springs, NY
    Period7/8/137/11/13

    Fingerprint

    Case based reasoning
    Availability
    Feedback
    Case-based Reasoning
    Domain Knowledge
    Violate
    Reuse
    Complement
    Clustering
    Configuration
    Computing
    Learning
    Collaboration
    Context

    Keywords

    • Case Alignment
    • Case Cohesion
    • Density Clustering
    • Multidisciplinary Collaboration
    • Recommender Systems
    • Single Class Learning
    • Subspace Clustering

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Gunawardena, S., Weber, R. O., & Stoyanovich, J. (2013). Learning feature weights from positive cases. In Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings (pp. 134-148). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7969 LNAI). https://doi.org/10.1007/978-3-642-39056-2_10

    Learning feature weights from positive cases. / Gunawardena, Sidath; Weber, Rosina O.; Stoyanovich, Julia.

    Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings. 2013. p. 134-148 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7969 LNAI).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Gunawardena, S, Weber, RO & Stoyanovich, J 2013, Learning feature weights from positive cases. in Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7969 LNAI, pp. 134-148, 21st International Conference on Case-Based Reasoning Research and Development, ICCBR 2013, Saratoga Springs, NY, United States, 7/8/13. https://doi.org/10.1007/978-3-642-39056-2_10
    Gunawardena S, Weber RO, Stoyanovich J. Learning feature weights from positive cases. In Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings. 2013. p. 134-148. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-39056-2_10
    Gunawardena, Sidath ; Weber, Rosina O. ; Stoyanovich, Julia. / Learning feature weights from positive cases. Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings. 2013. pp. 134-148 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{7d86b887f9ae4dea9a0223104fff5380,
    title = "Learning feature weights from positive cases",
    abstract = "The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse experiences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to recommend the characteristics of two collaborators to complement an academic seeking a multidisciplinary team; the three form a collaboration that resembles a configuration that has been successful in securing funding. While seeking a suitable measure for computing similarity between cases, we were confronted with two challenges: a problem context with insufficient domain knowledge and data that consists exclusively of successful collaborations, that is, it contains only positive instances. We present our strategy to overcome these challenges, which is a clustering-based approach to learn feature weights. Our approach identifies poorly aligned cases, i.e., ones that violate the assumption that similar problems have similar solutions. We use the poorly aligned cases as negatives in a feedback algorithm to learn feature weights. The result of this work is an integration of methods that makes CBR useful to yet another context and in conditions it has not been used before.",
    keywords = "Case Alignment, Case Cohesion, Density Clustering, Multidisciplinary Collaboration, Recommender Systems, Single Class Learning, Subspace Clustering",
    author = "Sidath Gunawardena and Weber, {Rosina O.} and Julia Stoyanovich",
    year = "2013",
    month = "9",
    day = "27",
    doi = "10.1007/978-3-642-39056-2_10",
    language = "English (US)",
    isbn = "9783642390555",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "134--148",
    booktitle = "Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings",

    }

    TY - GEN

    T1 - Learning feature weights from positive cases

    AU - Gunawardena, Sidath

    AU - Weber, Rosina O.

    AU - Stoyanovich, Julia

    PY - 2013/9/27

    Y1 - 2013/9/27

    N2 - The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse experiences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to recommend the characteristics of two collaborators to complement an academic seeking a multidisciplinary team; the three form a collaboration that resembles a configuration that has been successful in securing funding. While seeking a suitable measure for computing similarity between cases, we were confronted with two challenges: a problem context with insufficient domain knowledge and data that consists exclusively of successful collaborations, that is, it contains only positive instances. We present our strategy to overcome these challenges, which is a clustering-based approach to learn feature weights. Our approach identifies poorly aligned cases, i.e., ones that violate the assumption that similar problems have similar solutions. We use the poorly aligned cases as negatives in a feedback algorithm to learn feature weights. The result of this work is an integration of methods that makes CBR useful to yet another context and in conditions it has not been used before.

    AB - The availability of new data sources presents both opportunities and challenges for the use of Case-based Reasoning to solve novel problems. In this paper, we describe the research challenges we faced when trying to reuse experiences of successful academic collaborations available online in descriptions of funded grant proposals. The goal is to recommend the characteristics of two collaborators to complement an academic seeking a multidisciplinary team; the three form a collaboration that resembles a configuration that has been successful in securing funding. While seeking a suitable measure for computing similarity between cases, we were confronted with two challenges: a problem context with insufficient domain knowledge and data that consists exclusively of successful collaborations, that is, it contains only positive instances. We present our strategy to overcome these challenges, which is a clustering-based approach to learn feature weights. Our approach identifies poorly aligned cases, i.e., ones that violate the assumption that similar problems have similar solutions. We use the poorly aligned cases as negatives in a feedback algorithm to learn feature weights. The result of this work is an integration of methods that makes CBR useful to yet another context and in conditions it has not been used before.

    KW - Case Alignment

    KW - Case Cohesion

    KW - Density Clustering

    KW - Multidisciplinary Collaboration

    KW - Recommender Systems

    KW - Single Class Learning

    KW - Subspace Clustering

    UR - http://www.scopus.com/inward/record.url?scp=84884520840&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84884520840&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-39056-2_10

    DO - 10.1007/978-3-642-39056-2_10

    M3 - Conference contribution

    AN - SCOPUS:84884520840

    SN - 9783642390555

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 134

    EP - 148

    BT - Case-Based Reasoning Research and Development - 21st International Conference, ICCBR 2013, Proceedings

    ER -