A Caltech Library Service

Finding Social Media Trolls: Dynamic Keyword Selection Methods for Rapidly-Evolving Online Debates

Liu, Anqi and Srikanth, Maya and Adams-Cohen, Nicholas and Alvarez, R. Michael and Anandkumar, Anima (2019) Finding Social Media Trolls: Dynamic Keyword Selection Methods for Rapidly-Evolving Online Debates. . (Unpublished)

[img] PDF - Submitted Version
See Usage Policy.


Use this Persistent URL to link to this item:


Online harassment is a significant social problem. Prevention of online harassment requires rapid detection of harassing, offensive, and negative social media posts. In this paper, we propose the use of word embedding models to identify offensive and harassing social media messages in two aspects: detecting fast-changing topics for more effective data collection and representing word semantics in different domains. We demonstrate with preliminary results that using the GloVe (Global Vectors for Word Representation) model facilitates the discovery of new and relevant keywords to use for data collection and trolling detection. Our paper concludes with a discussion of a research agenda to further develop and test word embedding models for identification of social media harassment and trolling.

Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription Paper ItemCode InCaltech News
Adams-Cohen, Nicholas0000-0003-2251-1744
Alvarez, R. Michael0000-0002-8113-4451
Anandkumar, Anima0000-0002-6974-6797
Additional Information:Alvarez thanks the John Randolph Haynes and Dora Haynes for supporting his research in this area. Prof. Anandkumar is supported by Bren endowed Chair, faculty awards from Microsoft, Google, and Adobe, DARPA PAI and LwLL grants. Anqi Liu is a PIMCO postdoctoral fellow at Caltech. Codes Repository: The codes used for generating results in this paper can be accessed from the link:
Funding AgencyGrant Number
Bren Professor of Computing and Mathematical SciencesUNSPECIFIED
Defense Advanced Research Projects Agency (DARPA)UNSPECIFIED
Caltech PIMCO Graduate FellowshipUNSPECIFIED
Record Number:CaltechAUTHORS:20200108-154406622
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:100567
Deposited By: Tony Diaz
Deposited On:08 Jan 2020 23:54
Last Modified:02 Jun 2023 01:01

Repository Staff Only: item control page