Research Commons
      • Browse 
        • Communities & Collections
        • Titles
        • Authors
        • By Issue Date
        • Subjects
        • Types
        • Series
      • Help 
        • About
        • Collection Policy
        • OA Mandate Guidelines
        • Guidelines FAQ
        • Contact Us
      • My Account 
        • Sign In
        • Register
      View Item 
      •   Research Commons
      • University of Waikato Theses
      • Masters Degree Theses
      • View Item
      •   Research Commons
      • University of Waikato Theses
      • Masters Degree Theses
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Prompt-GAN – Customisable hate speech and extremist datasets via radicalised neural language models

      Govers, Jarod
      Thumbnail
      Files
      thesis.pdf
      8.747Mb
      Permanent link to Research Commons version
      https://hdl.handle.net/10289/15259
      Abstract
      Online hate speech and violent extremism knows no borders, no political boundaries, no remorse. Researchers face an uphill battle to collect hate speech data in volumes and topical diversity suitable for training state-of-the art content-moderation systems. Neural language models ushered in a new era of synthetic data generation in use across various businesses, all despite calls for research to protect against unintended toxic output. We present a method for radicalising pre-trained neural language models to identify real online hate speech, as well as present the risks of rouge radicalised AI bots which could undermine our trust in social media. We present Prompt-GAN, a prompt-tuning adversarial approach with three achievements. We demonstrate prompt-tuning’s ability to generate realistic types of hate and non-hate speech which mimics political extremist discourse. Prompt-GAN’s architecture offers a twofold reduction in memory and runtime requirements compared to fine-tuning. Finally, Prompt-GAN improves hate speech classification F1-scores by up to 10.1% and sets a new record in neural language simulation compared to the current state-of-the-art across three benchmark datasets.
      Date
      2022
      Type
      Thesis
      Degree Name
      Master of Cyber Security (MCS)
      Supervisors
      Patros, Panos
      Feldman, Philip
      Dant, Aaron
      Publisher
      The University of Waikato
      Rights
      All items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.
      Collections
      • Masters Degree Theses [2411]
      Show full item record  

      Usage

      Downloads, last 12 months
      263
       
       

      Usage Statistics

      For this itemFor all of Research Commons

      The University of Waikato - Te Whare Wānanga o WaikatoFeedback and RequestsCopyright and Legal Statement