Datasets for Statistical Quality Estimation for General Crowdsourcing Tasks
Grade label data
File format
- The 2nd to last columns of the first row contain the ID of each reviewer.
- The 2nd to last rows of the first column contain the ID of each artifact. The format of artifact ID is "[task ID]_[author ID]".
- The 2nd to last columns of each row contain the grade label of the artifact given by each reviewer.
- The grade labels are selected from 1(low) to 5(high).
Reference
If you use our dataset in your papers, please cite the following paper:
Yukino Baba and Hisashi Kashima: Statistical Quality Estimation for General Crowdsourcing Tasks, 19th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2013 (to appear).
Contact
Yukino Baba, The University of Tokyo.