Transformers, Humans in Disguise: Applying NLP Based Computational Techniques to Deliberative Democracy
Abstract/Contents
- Abstract
- This project seeks to apply a scalable, NLP based computational mechanism to automate portions of the work done by the Center for Deliberative Democracy (CDD) in their Deliberative Polling project. The machine based classifier that this project is attempting to profile is intended to emulate the currently manual task of tagging transcripts of worldwide deliberations about controversial topics with the intention of 1. Extracting argument dialogue segments and 2. Assessing argument deliberative quality as per the conventions set forth by the CDD. Top level, these techniques involve humans tagging transcripts using a 0, 1, 2, 2+ scale, reflecting the respective number of provided reasons to support an argument. The computational techniques that will be profiled for their capacity to replicate human tagging that this project will explore will be transformer based language models such as XLNet, a custom bag of words (BOW) + random-forest based classifier, and random chance. Empirical results have displayed an increase in model generalizability when transferring to the XLNet model, indicating a better capacity to emulate human transcript tagging across all categorizations, conversational subjects, and discussion locales.
Description
| Type of resource | text |
|---|---|
| Date modified | December 5, 2022 |
| Publication date | July 22, 2022 |
Creators/Contributors
| Author | Ahmed, Haroun |
|---|
Subjects
| Subject | Deliberative democracy |
|---|---|
| Subject | Natural language processing (Computer science) |
| Genre | Text |
| Genre | Thesis |
Bibliographic information
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Zero v1.0 Universal license (CC0).
Preferred citation
- Preferred citation
- Ahmed, H. (2022). Transformers, Humans in Disguise: Applying NLP Based Computational Techniques to Deliberative Democracy. Stanford Digital Repository. Available at https://purl.stanford.edu/nt256zn6858
Collection
Master's Theses, Symbolic Systems Program, Stanford University
View other items in this collection in SearchWorksContact information
- Contact
- harounahmed2@gmail.com
Also listed in
Loading usage metrics...