Answers are grounded in this paper and cite their sources.
The software development community, largely dominated by male developers, frequently exhibits instances of sexism and misogyny. While efforts have been made to automatically detect toxic and misogynistic language within this community, existing tools have yet to reach a satisfactory level of performance for practical implementation. However, there is growing interest in leveraging large language models (LLM) like ChatGPT for various tasks among developers. In this study, we aim to assess the effectiveness of ChatGPT for identifying particular forms of sexist remarks, such as maternal insult and stereotyping in software development communication channels. We experiment with various prompts, gradually scaffolding to fine-tune the task of detecting sexist remarks. Our preliminary analysis shows promising results in terms of identifying forms of sexist remarks such as maternal insult and stereotyping. Further research might consider investigating the dialogue as a unit of analysis to capture the nuances and context to better identify the remarks.
No takes yet. Share an insight, caveat, or question.
Sultana et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: