byPreetha Chatterjee · Slideshare

Preetha Chatterjee, profile picture

32 views

The document explores the effectiveness of OpenAI's ChatGPT in detecting toxic comments on GitHub, analyzing a dataset of 1,597 comments. Results indicate varying performance in identifying toxic behaviors and highlight specific types of toxicity that are often misclassified. It concludes that while ChatGPT is not perfect, it offers a viable method for automated toxicity detection without requiring extensive annotated datasets, with suggestions for further improvements and integration in software engineering workflows.

Related topics:

Download to read offline

Exploring ChatGPT for Toxicity Detection in GitHub Shyamal Mishra, Preetha Chatterjee 46th International Conference on Software Engineering (ICSE), April 2024 preetha.chatterjee@drexel.edu @PreethaChatterj https://preethac.github.io

Toxic Behaviors on GitHub 50% of OSS developers experienced negative interactions! (2017 GitHub Survey)

Research Questions • RQ1. How effective is OpenAI ChatGPT in detecting toxic text on GitHub? • RQ2. What types of toxic comments are misclassified by ChatGPT?

Dataset and Model 1597 GitHub issue comments (102 toxic, 1495 non-toxic) N.Raman, M.Cao, Y.Tsvetkov ,C.Kästner, and B.Vasilescu,“Stress and burnout in open source: Toward Finding, understanding, and mitigating unhealthy interactions,” ICSE-NIER ’20. GPT 3.5 Turbo

RQ1: How effective is OpenAI ChatGPT in detecting toxic text on GitHub? Prompt A Given the following software developer conversation, do you think this contains any toxicity? Please provide your answer in yes/no. Prompt B Examine whether the open-source developer community discussion includes toxicity. Very Toxic - a very hateful, aggressive, disrespectful comment or otherwise very likely to make a user leave a discussion or give up on sharing their perspective. Toxic - a rude, disrespectful, unreasonable comment or otherwise somewhat likely to make a user leave a discussion or give up on sharing their perspective. Slightly Toxic or hard to say. Not toxic. Temp Prompt Precision Recall F-Measure 0.2 Prompt A Prompt B 0.49 0.33 0.94 0.78 0.64 0.48 0.7 Prompt A Prompt B 0.43 0.39 0.49 0.71 0.55 0.51 1.2 Prompt A Prompt B 0.41 0.29 0.80 0.86 0.54 0.43

RQ2: What types of toxic comments are misclassified by ChatGPT? Error Category Count Labeling error Absence of Explicit Offensive Language Sarcasm and Irony Nuanced Toxicity Non-Responsive Answer Context-Dependent Toxicity Lengthy Phrasing 46 23 16 6 6 5 4 Few Examples … “Such a plugin already exists. And you chose to use a bleeding-edge build with it removed” “Thanks to you, I got ruined the world of survival and now have to do the cleaning map” “I can see it was a mistake trying to engage with you, so I’m locking this thread”

Discussion and Implications Model Interpretability Conversational Bot (Integration in SE Workflows) Prompt: Examine whether the software engineering community discussion includes toxicity. Please provide your answer in yes/no with a short justification. Statement: you’re useless, please leave ChatGPT Output: Yes. It includes personal attacks and a disrespectful tone. Statement: Pardon my English @friend, what you wrote is bullshit. ChatGPT Output: Yes. The use of the word "bullshit" is offensive, which can contribute to a toxic environment.

Summary Preprint: https://preethac.github.io/files/ICSE_NIER_24.pdf Dataset and replication package: https://anonymous.4open.science/r/open-source-toxicity-0236 ● Automated toxicity detection in software developer communication using a zero-shot LLM, namely ChatGPT, through several prompting approaches. ● While performance is not perfect, it eliminates the need for large annotated SE datasets. ● There are several avenues for improvement a. Exploring higher order models (e.g., GPT 4) and other LLMs (e.g., Gemini), b. Fine-tuning LLMs on data specific to software developer communities ● Bot integration holds particular significance in building trust among software engineers, encouraging their adoption and daily use in SE workflows

Read the original on slideshare.net ↗