A maximum entropy approach for collaborative filtering

John Browning, David Jonathan Miller

    Research output: Contribution to conference › Paper

    3 Citations (Scopus)

    Abstract

    Collaborative filtering (CF) involves predicting the preferences of a user for a set of items given partial knowledge of the user's preferences for other items, while leveraging a database of profiles for other users. CF has applications e.g. in predicting Web sites a person will visit and in automated searching of document databases. Fundamentally, CF is a pattern recognition task, but a formidable one, often involving a huge feature space, a large data set, and many missing features. Even more daunting is the fact that a CF inference engine must be capable of predicting any (user-selected) items, given any available set of partial knowledge on the user's other preferences. In other words, the model must be designed to solve any of a huge (combinatoric) set of possible inference tasks. CF techniques include memory-based, classification-based, and statistical modelling approaches. Among these, modelling approaches scale best with large data sets and are the most adept at handling missing features. The disadvantage of these methods lies in the statistical assumptions (e.g. feature independence), which may be unjustified. To address this shortcoming we propose a new model-based CF method, based on the maximum entropy principle. For the MS Web application, the new method is demonstrated to outperform a number of CF approaches, including naive Bayes and latent variable (cluster) models, support vector machines (SVMs), and the (Pearson) correlation method.

    Original languageEnglish (US)
    Pages3-12
    Number of pages10
    StatePublished - Dec 1 2001

    Fingerprint

    Collaborative filtering
    Entropy
    Inference engines
    Correlation methods
    Pattern recognition
    Support vector machines
    Websites
    Data storage equipment

    All Science Journal Classification (ASJC) codes

    • Signal Processing
    • Software
    • Electrical and Electronic Engineering

    Cite this

    Browning, John ; Miller, David Jonathan. / A maximum entropy approach for collaborative filtering. 10 p.
    @conference{91edf8abacdc470f90b523f31f48a0f4,
    title = "A maximum entropy approach for collaborative filtering",
    abstract = "Collaborative filtering (CF) involves predicting the preferences of a user for a set of items given partial knowledge of the user's preferences for other items, while leveraging a database of profiles for other users. CF has applications e.g. in predicting Web sites a person will visit and in automated searching of document databases. Fundamentally, CF is a pattern recognition task, but a formidable one, often involving a huge feature space, a large data set, and many missing features. Even more daunting is the fact that a CF inference engine must be capable of predicting any (user-selected) items, given any available set of partial knowledge on the user's other preferences. In other words, the model must be designed to solve any of a huge (combinatoric) set of possible inference tasks. CF techniques include memory-based, classification-based, and statistical modelling approaches. Among these, modelling approaches scale best with large data sets and are the most adept at handling missing features. The disadvantage of these methods lies in the statistical assumptions (e.g. feature independence), which may be unjustified. To address this shortcoming we propose a new model-based CF method, based on the maximum entropy principle. For the MS Web application, the new method is demonstrated to outperform a number of CF approaches, including naive Bayes and latent variable (cluster) models, support vector machines (SVMs), and the (Pearson) correlation method.",
    author = "John Browning and Miller, {David Jonathan}",
    year = "2001",
    month = "12",
    day = "1",
    language = "English (US)",
    pages = "3--12",

    }

    Browning, J & Miller, DJ 2001, 'A maximum entropy approach for collaborative filtering' pp. 3-12.

    A maximum entropy approach for collaborative filtering. / Browning, John; Miller, David Jonathan.

    2001. 3-12.

    Research output: Contribution to conference › Paper

    TY - CONF

    T1 - A maximum entropy approach for collaborative filtering

    AU - Browning, John

    AU - Miller, David Jonathan

    PY - 2001/12/1

    Y1 - 2001/12/1

    N2 - Collaborative filtering (CF) involves predicting the preferences of a user for a set of items given partial knowledge of the user's preferences for other items, while leveraging a database of profiles for other users. CF has applications e.g. in predicting Web sites a person will visit and in automated searching of document databases. Fundamentally, CF is a pattern recognition task, but a formidable one, often involving a huge feature space, a large data set, and many missing features. Even more daunting is the fact that a CF inference engine must be capable of predicting any (user-selected) items, given any available set of partial knowledge on the user's other preferences. In other words, the model must be designed to solve any of a huge (combinatoric) set of possible inference tasks. CF techniques include memory-based, classification-based, and statistical modelling approaches. Among these, modelling approaches scale best with large data sets and are the most adept at handling missing features. The disadvantage of these methods lies in the statistical assumptions (e.g. feature independence), which may be unjustified. To address this shortcoming we propose a new model-based CF method, based on the maximum entropy principle. For the MS Web application, the new method is demonstrated to outperform a number of CF approaches, including naive Bayes and latent variable (cluster) models, support vector machines (SVMs), and the (Pearson) correlation method.

    AB - Collaborative filtering (CF) involves predicting the preferences of a user for a set of items given partial knowledge of the user's preferences for other items, while leveraging a database of profiles for other users. CF has applications e.g. in predicting Web sites a person will visit and in automated searching of document databases. Fundamentally, CF is a pattern recognition task, but a formidable one, often involving a huge feature space, a large data set, and many missing features. Even more daunting is the fact that a CF inference engine must be capable of predicting any (user-selected) items, given any available set of partial knowledge on the user's other preferences. In other words, the model must be designed to solve any of a huge (combinatoric) set of possible inference tasks. CF techniques include memory-based, classification-based, and statistical modelling approaches. Among these, modelling approaches scale best with large data sets and are the most adept at handling missing features. The disadvantage of these methods lies in the statistical assumptions (e.g. feature independence), which may be unjustified. To address this shortcoming we propose a new model-based CF method, based on the maximum entropy principle. For the MS Web application, the new method is demonstrated to outperform a number of CF approaches, including naive Bayes and latent variable (cluster) models, support vector machines (SVMs), and the (Pearson) correlation method.

    UR - http://www.scopus.com/inward/record.url?scp=0035783845&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0035783845&partnerID=8YFLogxK

    M3 - Paper

    AN - SCOPUS:0035783845

    SP - 3

    EP - 12

    ER -

    Browning J, Miller DJ. A maximum entropy approach for collaborative filtering. 2001.

    Access to Document