Extracting social events based on timeline and user reliability analysis on twitter

    Research output: Contribution to conferenceConference paperpeer-review

    Abstract

    When some hot social issue or event occurs, it will significantly increase the number of comments and retweet on that day on twitter. Generally, an event can be extracted by its term frequency but it is hard to find an event that has a low term frequency. Because of this reason there can be a probability of missing important information. However, there is a kind of reliable user who is directly related to that event so that no matter how low the number of tweet is on that case. In this paper, we propose user reliability based event extraction method. The latent Dirichlet allocation(LDA) model is adapted with timeline analysis to extract high-frequency events. User behaviors are analyzed to classify reliable users who are directly related to the issue. Reliable low-frequency events can be detected based on reliable users. In order to verify the effectiveness of the proposed method, four social issues are selected and experimented on Korean twitter test set. The experimental results showed 97.2% in precision for the top 10 extracted events (P@10) on each day. This result shows that the proposed method is effective for extracting events in twitter corpus.

    Original languageEnglish
    Title of host publicationComputational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings
    PublisherSpringer Verlag
    Pages213-223
    Number of pages11
    EditionPART 2
    ISBN (Print)9783642549021
    DOIs
    StatePublished - 2014
    Event15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014 - Kathmandu, Nepal
    Duration: 2014.04.62014.04.12

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    NumberPART 2
    Volume8404 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014
    Country/TerritoryNepal
    CityKathmandu
    Period14.04.614.04.12

    Keywords

    • Event extraction
    • temporal LDA
    • timeline analysis
    • user behavior analysis

    Quacquarelli Symonds(QS) Subject Topics

    • Computer Science & Information Systems

    Fingerprint

    Dive into the research topics of 'Extracting social events based on timeline and user reliability analysis on twitter'. Together they form a unique fingerprint.

    Cite this