User-aware advertisability

Hai Tao Yu, Tetsuya Sakai

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

    Abstract

    In sponsored search, many studies focus on finding the most relevant advertisements (ads) and their optimal ranking for a submitted query. Determining whether it is suitable to show ads has received less attention. In this paper, we introduce the concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query. When computing the advertisability for a given query-user pair, we first classify the clicked web pages based on a pre-defined category hierarchy and use the aggregated topical categories of clicked web pages to represent user preference. Taking user preference into account, we then compute the ad-click probability for this query-user pair. Compared with existing methods, the experimental results show that user preference is of great value for generating user-specific advertisability. In particular, our approach that computes advertisability per query-user pair outperforms the two state-of-the-art methods that compute advertisability per query in terms of a variant of the normalized Discounted Cumulative Gain metric.

    Original languageEnglish
    Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Pages452-463
    Number of pages12
    Volume8281 LNCS
    DOIs
    Publication statusPublished - 2013
    Event9th Asia Information Retrieval Societies Conference on Information Retrieval Technology, AIRS 2013 - Singapore
    Duration: 2013 Dec 92013 Dec 11

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume8281 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other9th Asia Information Retrieval Societies Conference on Information Retrieval Technology, AIRS 2013
    CitySingapore
    Period13/12/913/12/11

    Fingerprint

    Websites
    Query
    User Preferences
    Ranking
    Classify
    Metric
    Computing
    Experimental Results

    ASJC Scopus subject areas

    • Computer Science(all)
    • Theoretical Computer Science

    Cite this

    Yu, H. T., & Sakai, T. (2013). User-aware advertisability. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8281 LNCS, pp. 452-463). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8281 LNCS). https://doi.org/10.1007/978-3-642-45068-6_39

    User-aware advertisability. / Yu, Hai Tao; Sakai, Tetsuya.

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8281 LNCS 2013. p. 452-463 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8281 LNCS).

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

    Yu, HT & Sakai, T 2013, User-aware advertisability. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8281 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8281 LNCS, pp. 452-463, 9th Asia Information Retrieval Societies Conference on Information Retrieval Technology, AIRS 2013, Singapore, 13/12/9. https://doi.org/10.1007/978-3-642-45068-6_39
    Yu HT, Sakai T. User-aware advertisability. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8281 LNCS. 2013. p. 452-463. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-45068-6_39
    Yu, Hai Tao ; Sakai, Tetsuya. / User-aware advertisability. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8281 LNCS 2013. pp. 452-463 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{241f5e589a3644ffbbd4fe4b0f833333,
    title = "User-aware advertisability",
    abstract = "In sponsored search, many studies focus on finding the most relevant advertisements (ads) and their optimal ranking for a submitted query. Determining whether it is suitable to show ads has received less attention. In this paper, we introduce the concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query. When computing the advertisability for a given query-user pair, we first classify the clicked web pages based on a pre-defined category hierarchy and use the aggregated topical categories of clicked web pages to represent user preference. Taking user preference into account, we then compute the ad-click probability for this query-user pair. Compared with existing methods, the experimental results show that user preference is of great value for generating user-specific advertisability. In particular, our approach that computes advertisability per query-user pair outperforms the two state-of-the-art methods that compute advertisability per query in terms of a variant of the normalized Discounted Cumulative Gain metric.",
    author = "Yu, {Hai Tao} and Tetsuya Sakai",
    year = "2013",
    doi = "10.1007/978-3-642-45068-6_39",
    language = "English",
    isbn = "9783642450679",
    volume = "8281 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    pages = "452--463",
    booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

    }

    TY - GEN

    T1 - User-aware advertisability

    AU - Yu, Hai Tao

    AU - Sakai, Tetsuya

    PY - 2013

    Y1 - 2013

    N2 - In sponsored search, many studies focus on finding the most relevant advertisements (ads) and their optimal ranking for a submitted query. Determining whether it is suitable to show ads has received less attention. In this paper, we introduce the concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query. When computing the advertisability for a given query-user pair, we first classify the clicked web pages based on a pre-defined category hierarchy and use the aggregated topical categories of clicked web pages to represent user preference. Taking user preference into account, we then compute the ad-click probability for this query-user pair. Compared with existing methods, the experimental results show that user preference is of great value for generating user-specific advertisability. In particular, our approach that computes advertisability per query-user pair outperforms the two state-of-the-art methods that compute advertisability per query in terms of a variant of the normalized Discounted Cumulative Gain metric.

    AB - In sponsored search, many studies focus on finding the most relevant advertisements (ads) and their optimal ranking for a submitted query. Determining whether it is suitable to show ads has received less attention. In this paper, we introduce the concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query. When computing the advertisability for a given query-user pair, we first classify the clicked web pages based on a pre-defined category hierarchy and use the aggregated topical categories of clicked web pages to represent user preference. Taking user preference into account, we then compute the ad-click probability for this query-user pair. Compared with existing methods, the experimental results show that user preference is of great value for generating user-specific advertisability. In particular, our approach that computes advertisability per query-user pair outperforms the two state-of-the-art methods that compute advertisability per query in terms of a variant of the normalized Discounted Cumulative Gain metric.

    UR - http://www.scopus.com/inward/record.url?scp=84893319629&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84893319629&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-45068-6_39

    DO - 10.1007/978-3-642-45068-6_39

    M3 - Conference contribution

    AN - SCOPUS:84893319629

    SN - 9783642450679

    VL - 8281 LNCS

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

    SP - 452

    EP - 463

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

    ER -