Understanding local structure in ranked datasets

Julia Stoyanovich, Sihem Amer-Yahia, Susan B. Davidson, Marie Jacob, Tova Milo

    Research output: Contribution to conferencePaper

    Abstract

    Ranked data is ubiquitous in real-world applications. Rankings arise naturally when users express preferences about products and services, when voters cast ballots in elections, when funding proposals are evaluated based on their merits and university departments based on their reputation, or when genes are ordered based on their expression levels under various experimental conditions. We observe that ranked data exhibits interesting local structure, representing agreement of subsets of rankers over subsets of items. Being able to model, identify and describe such structure is important, because it enables novel kinds of analysis with the potential of making ground-breaking impact, but is challenging to do effectively and efficiently. We argue for the use of fundamental data management principles such as declarativeness and incremental evaluation, in combination with state-of-the-art machine learning and data mining techniques, for addressing the effectiveness and efficiency challenges. We describe the key ingredients of a solution, and propose a roadmap towards a framework that will enable robust and efficient analysis of large ranked datasets.

    Original languageEnglish (US)
    StatePublished - Jan 1 2013
    Event6th Biennial Conference on Innovative Data Systems Research, CIDR 2013 - Pacific Grove, United States
    Duration: Jan 6 2013Jan 9 2013

    Conference

    Conference6th Biennial Conference on Innovative Data Systems Research, CIDR 2013
    CountryUnited States
    CityPacific Grove
    Period1/6/131/9/13

    Fingerprint

    Information management
    Data mining
    Learning systems
    Genes
    Data management
    User preferences
    Evaluation
    Gene
    Funding
    Ranking
    Machine learning
    Elections
    Voters
    Roadmap
    Incremental

    ASJC Scopus subject areas

    • Hardware and Architecture
    • Information Systems and Management
    • Artificial Intelligence
    • Information Systems

    Cite this

    Stoyanovich, J., Amer-Yahia, S., Davidson, S. B., Jacob, M., & Milo, T. (2013). Understanding local structure in ranked datasets. Paper presented at 6th Biennial Conference on Innovative Data Systems Research, CIDR 2013, Pacific Grove, United States.

    Understanding local structure in ranked datasets. / Stoyanovich, Julia; Amer-Yahia, Sihem; Davidson, Susan B.; Jacob, Marie; Milo, Tova.

    2013. Paper presented at 6th Biennial Conference on Innovative Data Systems Research, CIDR 2013, Pacific Grove, United States.

    Research output: Contribution to conferencePaper

    Stoyanovich, J, Amer-Yahia, S, Davidson, SB, Jacob, M & Milo, T 2013, 'Understanding local structure in ranked datasets', Paper presented at 6th Biennial Conference on Innovative Data Systems Research, CIDR 2013, Pacific Grove, United States, 1/6/13 - 1/9/13.
    Stoyanovich J, Amer-Yahia S, Davidson SB, Jacob M, Milo T. Understanding local structure in ranked datasets. 2013. Paper presented at 6th Biennial Conference on Innovative Data Systems Research, CIDR 2013, Pacific Grove, United States.
    Stoyanovich, Julia ; Amer-Yahia, Sihem ; Davidson, Susan B. ; Jacob, Marie ; Milo, Tova. / Understanding local structure in ranked datasets. Paper presented at 6th Biennial Conference on Innovative Data Systems Research, CIDR 2013, Pacific Grove, United States.
    @conference{af13ffb68d5844988ad790529d43dc40,
    title = "Understanding local structure in ranked datasets",
    abstract = "Ranked data is ubiquitous in real-world applications. Rankings arise naturally when users express preferences about products and services, when voters cast ballots in elections, when funding proposals are evaluated based on their merits and university departments based on their reputation, or when genes are ordered based on their expression levels under various experimental conditions. We observe that ranked data exhibits interesting local structure, representing agreement of subsets of rankers over subsets of items. Being able to model, identify and describe such structure is important, because it enables novel kinds of analysis with the potential of making ground-breaking impact, but is challenging to do effectively and efficiently. We argue for the use of fundamental data management principles such as declarativeness and incremental evaluation, in combination with state-of-the-art machine learning and data mining techniques, for addressing the effectiveness and efficiency challenges. We describe the key ingredients of a solution, and propose a roadmap towards a framework that will enable robust and efficient analysis of large ranked datasets.",
    author = "Julia Stoyanovich and Sihem Amer-Yahia and Davidson, {Susan B.} and Marie Jacob and Tova Milo",
    year = "2013",
    month = "1",
    day = "1",
    language = "English (US)",
    note = "6th Biennial Conference on Innovative Data Systems Research, CIDR 2013 ; Conference date: 06-01-2013 Through 09-01-2013",

    }

    TY - CONF

    T1 - Understanding local structure in ranked datasets

    AU - Stoyanovich, Julia

    AU - Amer-Yahia, Sihem

    AU - Davidson, Susan B.

    AU - Jacob, Marie

    AU - Milo, Tova

    PY - 2013/1/1

    Y1 - 2013/1/1

    N2 - Ranked data is ubiquitous in real-world applications. Rankings arise naturally when users express preferences about products and services, when voters cast ballots in elections, when funding proposals are evaluated based on their merits and university departments based on their reputation, or when genes are ordered based on their expression levels under various experimental conditions. We observe that ranked data exhibits interesting local structure, representing agreement of subsets of rankers over subsets of items. Being able to model, identify and describe such structure is important, because it enables novel kinds of analysis with the potential of making ground-breaking impact, but is challenging to do effectively and efficiently. We argue for the use of fundamental data management principles such as declarativeness and incremental evaluation, in combination with state-of-the-art machine learning and data mining techniques, for addressing the effectiveness and efficiency challenges. We describe the key ingredients of a solution, and propose a roadmap towards a framework that will enable robust and efficient analysis of large ranked datasets.

    AB - Ranked data is ubiquitous in real-world applications. Rankings arise naturally when users express preferences about products and services, when voters cast ballots in elections, when funding proposals are evaluated based on their merits and university departments based on their reputation, or when genes are ordered based on their expression levels under various experimental conditions. We observe that ranked data exhibits interesting local structure, representing agreement of subsets of rankers over subsets of items. Being able to model, identify and describe such structure is important, because it enables novel kinds of analysis with the potential of making ground-breaking impact, but is challenging to do effectively and efficiently. We argue for the use of fundamental data management principles such as declarativeness and incremental evaluation, in combination with state-of-the-art machine learning and data mining techniques, for addressing the effectiveness and efficiency challenges. We describe the key ingredients of a solution, and propose a roadmap towards a framework that will enable robust and efficient analysis of large ranked datasets.

    UR - http://www.scopus.com/inward/record.url?scp=84905118362&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84905118362&partnerID=8YFLogxK

    M3 - Paper

    AN - SCOPUS:84905118362

    ER -