Evaluating the performance and privacy of a token-based collaborative recommender

Ville Ollikainen, Valtteri Niemi

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    1 Citation (Scopus)

    Abstract

    The rapid expansion of available online services has raised concerns about user privacy. In the online world, only a minority of users is actually aware where their data is stored and the policies, how the data may be eventually used. However, at the same time consumers expect more quality from online services, demanding personalized services that fit their individual needs, preferences and values. One approach for service personalization is to use collaborative recommenders. From the privacy perspective, mainstream collaborative recommenders present an inherent security risk, since they are based on memorizing useritem transactions. In this paper, we will study a recently developed token-based method (sometimes referred as an acronym "upcv") which creates privacy-protecting abstraction that is based on collections of randomly generated tokens. These collections are capable of providing information for collaborative recommendations without maintaining any transactional history. This paper presents quality evaluation of item-To-item recommendations using the token-based collaborative recommender, utilizing ISBN agencies of Book-Crossing dataset (BX) books at the data set. This paper will also discuss challenges related to BX. Privacy issues are evaluated with a specific emphasis on the concept of deniability.
    Original languageEnglish
    Title of host publicationProceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
    PublisherAssociation for Computing Machinery ACM
    Pages1049-1053
    Number of pages5
    ISBN (Electronic)9781450349512
    ISBN (Print)978-1-4503-4951-2
    DOIs
    Publication statusPublished - 23 Aug 2017
    MoE publication typeA4 Article in a conference publication
    Event16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017 - Leipzig, Germany
    Duration: 23 Aug 201726 Aug 2017

    Conference

    Conference16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017
    Abbreviated titleWI 2017
    CountryGermany
    CityLeipzig
    Period23/08/1726/08/17

    Keywords

    • book-crossing
    • collaborative recommender
    • deniability
    • privacy
    • token
    • upcv

    Cite this

    Ollikainen, V., & Niemi, V. (2017). Evaluating the performance and privacy of a token-based collaborative recommender. In Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017 (pp. 1049-1053). Association for Computing Machinery ACM. https://doi.org/10.1145/3106426.3109434
    Ollikainen, Ville ; Niemi, Valtteri. / Evaluating the performance and privacy of a token-based collaborative recommender. Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017. Association for Computing Machinery ACM, 2017. pp. 1049-1053
    @inproceedings{c697681f64554ba0802a0b8e420b1773,
    title = "Evaluating the performance and privacy of a token-based collaborative recommender",
    abstract = "The rapid expansion of available online services has raised concerns about user privacy. In the online world, only a minority of users is actually aware where their data is stored and the policies, how the data may be eventually used. However, at the same time consumers expect more quality from online services, demanding personalized services that fit their individual needs, preferences and values. One approach for service personalization is to use collaborative recommenders. From the privacy perspective, mainstream collaborative recommenders present an inherent security risk, since they are based on memorizing useritem transactions. In this paper, we will study a recently developed token-based method (sometimes referred as an acronym {"}upcv{"}) which creates privacy-protecting abstraction that is based on collections of randomly generated tokens. These collections are capable of providing information for collaborative recommendations without maintaining any transactional history. This paper presents quality evaluation of item-To-item recommendations using the token-based collaborative recommender, utilizing ISBN agencies of Book-Crossing dataset (BX) books at the data set. This paper will also discuss challenges related to BX. Privacy issues are evaluated with a specific emphasis on the concept of deniability.",
    keywords = "book-crossing, collaborative recommender, deniability, privacy, token, upcv",
    author = "Ville Ollikainen and Valtteri Niemi",
    note = "CA2: BA1615",
    year = "2017",
    month = "8",
    day = "23",
    doi = "10.1145/3106426.3109434",
    language = "English",
    isbn = "978-1-4503-4951-2",
    pages = "1049--1053",
    booktitle = "Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017",
    publisher = "Association for Computing Machinery ACM",
    address = "United States",

    }

    Ollikainen, V & Niemi, V 2017, Evaluating the performance and privacy of a token-based collaborative recommender. in Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017. Association for Computing Machinery ACM, pp. 1049-1053, 16th IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017, Leipzig, Germany, 23/08/17. https://doi.org/10.1145/3106426.3109434

    Evaluating the performance and privacy of a token-based collaborative recommender. / Ollikainen, Ville; Niemi, Valtteri.

    Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017. Association for Computing Machinery ACM, 2017. p. 1049-1053.

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    TY - GEN

    T1 - Evaluating the performance and privacy of a token-based collaborative recommender

    AU - Ollikainen, Ville

    AU - Niemi, Valtteri

    N1 - CA2: BA1615

    PY - 2017/8/23

    Y1 - 2017/8/23

    N2 - The rapid expansion of available online services has raised concerns about user privacy. In the online world, only a minority of users is actually aware where their data is stored and the policies, how the data may be eventually used. However, at the same time consumers expect more quality from online services, demanding personalized services that fit their individual needs, preferences and values. One approach for service personalization is to use collaborative recommenders. From the privacy perspective, mainstream collaborative recommenders present an inherent security risk, since they are based on memorizing useritem transactions. In this paper, we will study a recently developed token-based method (sometimes referred as an acronym "upcv") which creates privacy-protecting abstraction that is based on collections of randomly generated tokens. These collections are capable of providing information for collaborative recommendations without maintaining any transactional history. This paper presents quality evaluation of item-To-item recommendations using the token-based collaborative recommender, utilizing ISBN agencies of Book-Crossing dataset (BX) books at the data set. This paper will also discuss challenges related to BX. Privacy issues are evaluated with a specific emphasis on the concept of deniability.

    AB - The rapid expansion of available online services has raised concerns about user privacy. In the online world, only a minority of users is actually aware where their data is stored and the policies, how the data may be eventually used. However, at the same time consumers expect more quality from online services, demanding personalized services that fit their individual needs, preferences and values. One approach for service personalization is to use collaborative recommenders. From the privacy perspective, mainstream collaborative recommenders present an inherent security risk, since they are based on memorizing useritem transactions. In this paper, we will study a recently developed token-based method (sometimes referred as an acronym "upcv") which creates privacy-protecting abstraction that is based on collections of randomly generated tokens. These collections are capable of providing information for collaborative recommendations without maintaining any transactional history. This paper presents quality evaluation of item-To-item recommendations using the token-based collaborative recommender, utilizing ISBN agencies of Book-Crossing dataset (BX) books at the data set. This paper will also discuss challenges related to BX. Privacy issues are evaluated with a specific emphasis on the concept of deniability.

    KW - book-crossing

    KW - collaborative recommender

    KW - deniability

    KW - privacy

    KW - token

    KW - upcv

    UR - http://www.scopus.com/inward/record.url?scp=85030982065&partnerID=8YFLogxK

    U2 - 10.1145/3106426.3109434

    DO - 10.1145/3106426.3109434

    M3 - Conference article in proceedings

    SN - 978-1-4503-4951-2

    SP - 1049

    EP - 1053

    BT - Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017

    PB - Association for Computing Machinery ACM

    ER -

    Ollikainen V, Niemi V. Evaluating the performance and privacy of a token-based collaborative recommender. In Proceedings - 2017 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2017. Association for Computing Machinery ACM. 2017. p. 1049-1053 https://doi.org/10.1145/3106426.3109434