Personalized fitting with deviation adjustment based on support vector regression for recommendation

Weimin Li, Mengke Yao, Qun Jin

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

    Abstract

    Almost all of the existing Collaborative Filtering (CF) methods rely only on the rating data while ignoring some important implicit information in non-rating properties for users and items, which have a significant impact on the preference. In this study, considering that the average rating of users and items have a certain stability, we firstly propose a personalized fitting pattern to predict missing ratings based on the trusty score set, which combines both the user-based CF and item-based CF. In order to further reduce the prediction error, we use the non-rating attributes, such as a user’s age, gender and occupation, and an item’s release date and price. Moreover, we present the deviation adjustment method based on the Support Vector Regression (SVR). Experiment results show that our proposed algorithms can increase the accuracy of recommendation versus the traditional CF.

    Original languageEnglish
    Title of host publicationLecture Notes in Electrical Engineering
    PublisherSpringer Verlag
    Pages173-178
    Number of pages6
    Volume308
    ISBN (Print)9783642548994
    DOIs
    Publication statusPublished - 2014
    EventFTRA 8th International Conference on Multimedia and Ubiquitous Engineering, MUE 2014 - Zhangjiajie
    Duration: 2014 May 282014 May 31

    Publication series

    NameLecture Notes in Electrical Engineering
    Volume308
    ISSN (Print)18761100
    ISSN (Electronic)18761119

    Other

    OtherFTRA 8th International Conference on Multimedia and Ubiquitous Engineering, MUE 2014
    CityZhangjiajie
    Period14/5/2814/5/31

    Fingerprint

    Collaborative filtering
    Experiments

    Keywords

    • Collaborative filtering
    • Deviation adjustment
    • Personalized fitting
    • Recommendation

    ASJC Scopus subject areas

    • Industrial and Manufacturing Engineering

    Cite this

    Li, W., Yao, M., & Jin, Q. (2014). Personalized fitting with deviation adjustment based on support vector regression for recommendation. In Lecture Notes in Electrical Engineering (Vol. 308, pp. 173-178). (Lecture Notes in Electrical Engineering; Vol. 308). Springer Verlag. https://doi.org/10.1007/978-3-642-54900-7_25

    Personalized fitting with deviation adjustment based on support vector regression for recommendation. / Li, Weimin; Yao, Mengke; Jin, Qun.

    Lecture Notes in Electrical Engineering. Vol. 308 Springer Verlag, 2014. p. 173-178 (Lecture Notes in Electrical Engineering; Vol. 308).

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

    Li, W, Yao, M & Jin, Q 2014, Personalized fitting with deviation adjustment based on support vector regression for recommendation. in Lecture Notes in Electrical Engineering. vol. 308, Lecture Notes in Electrical Engineering, vol. 308, Springer Verlag, pp. 173-178, FTRA 8th International Conference on Multimedia and Ubiquitous Engineering, MUE 2014, Zhangjiajie, 14/5/28. https://doi.org/10.1007/978-3-642-54900-7_25
    Li W, Yao M, Jin Q. Personalized fitting with deviation adjustment based on support vector regression for recommendation. In Lecture Notes in Electrical Engineering. Vol. 308. Springer Verlag. 2014. p. 173-178. (Lecture Notes in Electrical Engineering). https://doi.org/10.1007/978-3-642-54900-7_25
    Li, Weimin ; Yao, Mengke ; Jin, Qun. / Personalized fitting with deviation adjustment based on support vector regression for recommendation. Lecture Notes in Electrical Engineering. Vol. 308 Springer Verlag, 2014. pp. 173-178 (Lecture Notes in Electrical Engineering).
    @inproceedings{f99080fa89294156bae54916ee594e38,
    title = "Personalized fitting with deviation adjustment based on support vector regression for recommendation",
    abstract = "Almost all of the existing Collaborative Filtering (CF) methods rely only on the rating data while ignoring some important implicit information in non-rating properties for users and items, which have a significant impact on the preference. In this study, considering that the average rating of users and items have a certain stability, we firstly propose a personalized fitting pattern to predict missing ratings based on the trusty score set, which combines both the user-based CF and item-based CF. In order to further reduce the prediction error, we use the non-rating attributes, such as a user’s age, gender and occupation, and an item’s release date and price. Moreover, we present the deviation adjustment method based on the Support Vector Regression (SVR). Experiment results show that our proposed algorithms can increase the accuracy of recommendation versus the traditional CF.",
    keywords = "Collaborative filtering, Deviation adjustment, Personalized fitting, Recommendation",
    author = "Weimin Li and Mengke Yao and Qun Jin",
    year = "2014",
    doi = "10.1007/978-3-642-54900-7_25",
    language = "English",
    isbn = "9783642548994",
    volume = "308",
    series = "Lecture Notes in Electrical Engineering",
    publisher = "Springer Verlag",
    pages = "173--178",
    booktitle = "Lecture Notes in Electrical Engineering",

    }

    TY - GEN

    T1 - Personalized fitting with deviation adjustment based on support vector regression for recommendation

    AU - Li, Weimin

    AU - Yao, Mengke

    AU - Jin, Qun

    PY - 2014

    Y1 - 2014

    N2 - Almost all of the existing Collaborative Filtering (CF) methods rely only on the rating data while ignoring some important implicit information in non-rating properties for users and items, which have a significant impact on the preference. In this study, considering that the average rating of users and items have a certain stability, we firstly propose a personalized fitting pattern to predict missing ratings based on the trusty score set, which combines both the user-based CF and item-based CF. In order to further reduce the prediction error, we use the non-rating attributes, such as a user’s age, gender and occupation, and an item’s release date and price. Moreover, we present the deviation adjustment method based on the Support Vector Regression (SVR). Experiment results show that our proposed algorithms can increase the accuracy of recommendation versus the traditional CF.

    AB - Almost all of the existing Collaborative Filtering (CF) methods rely only on the rating data while ignoring some important implicit information in non-rating properties for users and items, which have a significant impact on the preference. In this study, considering that the average rating of users and items have a certain stability, we firstly propose a personalized fitting pattern to predict missing ratings based on the trusty score set, which combines both the user-based CF and item-based CF. In order to further reduce the prediction error, we use the non-rating attributes, such as a user’s age, gender and occupation, and an item’s release date and price. Moreover, we present the deviation adjustment method based on the Support Vector Regression (SVR). Experiment results show that our proposed algorithms can increase the accuracy of recommendation versus the traditional CF.

    KW - Collaborative filtering

    KW - Deviation adjustment

    KW - Personalized fitting

    KW - Recommendation

    UR - http://www.scopus.com/inward/record.url?scp=84924345950&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84924345950&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-54900-7_25

    DO - 10.1007/978-3-642-54900-7_25

    M3 - Conference contribution

    AN - SCOPUS:84924345950

    SN - 9783642548994

    VL - 308

    T3 - Lecture Notes in Electrical Engineering

    SP - 173

    EP - 178

    BT - Lecture Notes in Electrical Engineering

    PB - Springer Verlag

    ER -