IDEAS home Printed from https://ideas.repec.org/h/spr/sprchp/978-981-99-0803-5_3.html

Exclusive Topic Model

In: Research Papers in Statistical Inference for Time Series and Related Models

Author

Listed:
  • Hao Lei

    (National University of Singapore)

  • Kailiang Liu

    (National University of Singapore)

  • Ying Chen

    (National University of Singapore)

Abstract

Digital documents are generated, disseminated, and disclosed in books, research papers, newspapers, online feedback, and other content containing large amounts of information, for which discovering topics becomes important but challenging. In the current field of topic modeling, there are limited techniques available to deal with (1) the predominance of frequently occurring words in the estimated topics and (2) the overlap of commonly used words in different topics. We propose exclusive topic modeling (ETM) to identify field-specific keywords that typically occur less frequently but are important for representing certain topics, and to provide well-structured exclusive terms for topics within each topic. Specifically, we impose a weighted LASSO penalty to automatically reduce the dominance of frequently occurring but less relevant words and a pairwise Kullback–Leibler divergence penalty to achieve topic separation. Numerical studies show that the ETM can detect field-specific keywords and provide exclusive topics, which is more meaningful for interpretation and topic detection than other models such as the latent Dirichlet allocation model.

Suggested Citation

  • Hao Lei & Kailiang Liu & Ying Chen, 2023. "Exclusive Topic Model," Springer Books, in: Yan Liu & Junichi Hirukawa & Yoshihide Kakizawa (ed.), Research Papers in Statistical Inference for Time Series and Related Models, chapter 0, pages 83-109, Springer.
    • Handle: RePEc:spr:sprchp:978-981-99-0803-5_3
    DOI: 10.1007/978-981-99-0803-5_3
    as

    Download full text from publisher

    To our knowledge, this item is not available for download. To find whether it is available, there are three options:
    1. Check below whether another version of this item is available online.
    2. Check on the provider's web page whether it is in fact available.
    3. Perform a
    for a similarly titled item that would be available.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:spr:sprchp:978-981-99-0803-5_3. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    We have no bibliographic references for this item. You can help adding them by using this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.springer.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.