Toward a Robust and Universal Crowd-Labeling Framework

Abstract

One of the main challenges in crowd-labeling is to control for or determine in advance the proportion of low-quality/malicious labelers. We propose methods that estimate the labeler and data instance related parameters using frequentist and Bayesian approaches. All these approaches are based on expert-labeled instance (ground truth) for a small percentage of data to learn the parameters. We also derive a lower bound on the number of expert-labeled instances needed to get better quality labels. PDF

Cite

Text

Khattak. "Toward a Robust and Universal Crowd-Labeling Framework." International Joint Conference on Artificial Intelligence, 2016.

Markdown

[Khattak. "Toward a Robust and Universal Crowd-Labeling Framework." International Joint Conference on Artificial Intelligence, 2016.](https://mlanthology.org/ijcai/2016/khattak2016ijcai-robust/)

BibTeX

@inproceedings{khattak2016ijcai-robust,
  title     = {{Toward a Robust and Universal Crowd-Labeling Framework}},
  author    = {Khattak, Faiza Khan},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2016},
  pages     = {4006-4007},
  url       = {https://mlanthology.org/ijcai/2016/khattak2016ijcai-robust/}
}