App2Vec: Vector modeling of mobile apps and applications

Qiang Ma, Shanmugavelayutham Muthukrishnan, Wil Simpson

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

    Abstract

    We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82% of the retrieved top similar apps are semantically relevant, achieving 37% lift over bag-of-word approach and 140% lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5%. This is the first industry scale design, training and use of app vectorization.

    Original languageEnglish (US)
    Title of host publicationProceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
    EditorsRavi Kumar, James Caverlee, Hanghang Tong
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages599-606
    Number of pages8
    ISBN (Electronic)9781509028467
    DOIs
    StatePublished - Nov 21 2016
    Event2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016 - San Francisco, United States
    Duration: Aug 18 2016Aug 21 2016

    Publication series

    NameProceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

    Conference

    Conference2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016
    CountryUnited States
    CitySan Francisco
    Period8/18/168/21/16

    Fingerprint

    Application programs
    time series
    semantics
    industry
    evaluation
    learning
    Factorization
    Time series
    Semantics
    time

    ASJC Scopus subject areas

    • Computer Networks and Communications
    • Sociology and Political Science
    • Communication

    Cite this

    Ma, Q., Muthukrishnan, S., & Simpson, W. (2016). App2Vec: Vector modeling of mobile apps and applications. In R. Kumar, J. Caverlee, & H. Tong (Eds.), Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016 (pp. 599-606). [7752297] (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ASONAM.2016.7752297

    App2Vec : Vector modeling of mobile apps and applications. / Ma, Qiang; Muthukrishnan, Shanmugavelayutham; Simpson, Wil.

    Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. ed. / Ravi Kumar; James Caverlee; Hanghang Tong. Institute of Electrical and Electronics Engineers Inc., 2016. p. 599-606 7752297 (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016).

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

    Ma, Q, Muthukrishnan, S & Simpson, W 2016, App2Vec: Vector modeling of mobile apps and applications. in R Kumar, J Caverlee & H Tong (eds), Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016., 7752297, Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016, Institute of Electrical and Electronics Engineers Inc., pp. 599-606, 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016, San Francisco, United States, 8/18/16. https://doi.org/10.1109/ASONAM.2016.7752297
    Ma Q, Muthukrishnan S, Simpson W. App2Vec: Vector modeling of mobile apps and applications. In Kumar R, Caverlee J, Tong H, editors, Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 599-606. 7752297. (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016). https://doi.org/10.1109/ASONAM.2016.7752297
    Ma, Qiang ; Muthukrishnan, Shanmugavelayutham ; Simpson, Wil. / App2Vec : Vector modeling of mobile apps and applications. Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016. editor / Ravi Kumar ; James Caverlee ; Hanghang Tong. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 599-606 (Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016).
    @inproceedings{c56b2f458b7947de97bd65de5582fb63,
    title = "App2Vec: Vector modeling of mobile apps and applications",
    abstract = "We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82{\%} of the retrieved top similar apps are semantically relevant, achieving 37{\%} lift over bag-of-word approach and 140{\%} lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5{\%}. This is the first industry scale design, training and use of app vectorization.",
    author = "Qiang Ma and Shanmugavelayutham Muthukrishnan and Wil Simpson",
    year = "2016",
    month = "11",
    day = "21",
    doi = "10.1109/ASONAM.2016.7752297",
    language = "English (US)",
    series = "Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    pages = "599--606",
    editor = "Ravi Kumar and James Caverlee and Hanghang Tong",
    booktitle = "Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016",

    }

    TY - GEN

    T1 - App2Vec

    T2 - Vector modeling of mobile apps and applications

    AU - Ma, Qiang

    AU - Muthukrishnan, Shanmugavelayutham

    AU - Simpson, Wil

    PY - 2016/11/21

    Y1 - 2016/11/21

    N2 - We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82% of the retrieved top similar apps are semantically relevant, achieving 37% lift over bag-of-word approach and 140% lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5%. This is the first industry scale design, training and use of app vectorization.

    AB - We design a way to model apps as vectors, inspired by the recent deep learning approach to vectorization of words called word2vec. Our method relies on how users use apps. In particular, we visualize the time series of how each user uses mobile apps as a 'document', and apply the recent word2vec modeling on these documents, but the novelty is that the training context is carefully weighted by the time interval between the usage of successive apps. This gives us the app2vec vectorization of apps. We apply this to industrial scale data from Yahoo! and (a) show examples that app2vec captures semantic relationships between apps, much as word2vec does with words, (b) show using Yahoo!'s extensive human evaluation system that 82% of the retrieved top similar apps are semantically relevant, achieving 37% lift over bag-of-word approach and 140% lift over matrix factorization approach to vectorizing apps, and (c) finally, we use app2vec to predict app-install conversion and improve ad conversion prediction accuracy by almost 5%. This is the first industry scale design, training and use of app vectorization.

    UR - http://www.scopus.com/inward/record.url?scp=85006810397&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85006810397&partnerID=8YFLogxK

    U2 - 10.1109/ASONAM.2016.7752297

    DO - 10.1109/ASONAM.2016.7752297

    M3 - Conference contribution

    AN - SCOPUS:85006810397

    T3 - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

    SP - 599

    EP - 606

    BT - Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016

    A2 - Kumar, Ravi

    A2 - Caverlee, James

    A2 - Tong, Hanghang

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -