This is a dataset released from the paper Understanding Users’ Dissatisfaction with ChatGPT Responses: Types, Resolving Tactics, and the Effect of Knowledge Level.
It comprises instances where users felt dissatisfied with responses from ChatGPT during the conversations and how they responded to them. It covers the aspects in which users felt dissatisfaction with the ChatGPT's response, the severity of the dissatisfaction, how users responded to these dissatisfactory responses in their subsequent prompts, and the effectiveness of the users' tactic employed in their prompts.
The dataset is hierarchically organized, comprising the following components:
- User (N=94)
- ChatGPT conversation links and logs (N=249)
- User's recollected experience data on dissatisfactory ChatGPT responses (N=377)
- User's strategies to respond to the dissatisfactory response (N=459)
A detailed description of the dataset is here:
https://github.com/kixlab/chatgpt-dataset