Skip to content

Data quality metrics #71

Description

@faneshion

In this issue, we discuss the potential metric used to evaluate the quality of an input dataset. The quality of dataset is very important since there are many automatically generated dataset built on ChatGPT.

Some potential quality dimensions related to a QA dataset may be:
1) the number of instances;
2) the diversity of the question topic;
3) the length of the answer;
4) the fluency of the answer;
5) the correctness of the answer;
6) the length of contexts;
7) the relationship between the context and the answer to a question.

we can dicuss it in this issue.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions