We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: OpinionRank: Extracting Ground Truth Labels from Unreliable Expert Opinions with Graph-Based Spectral Ranking

Abstract: As larger and more comprehensive datasets become standard in contemporary machine learning, it becomes increasingly more difficult to obtain reliable, trustworthy label information with which to train sophisticated models. To address this problem, crowdsourcing has emerged as a popular, inexpensive, and efficient data mining solution for performing distributed label collection. However, crowdsourced annotations are inherently untrustworthy, as the labels are provided by anonymous volunteers who may have varying, unreliable expertise. Worse yet, some participants on commonly used platforms such as Amazon Mechanical Turk may be adversarial, and provide intentionally incorrect label information without the end user's knowledge. We discuss three conventional models of the label generation process, describing their parameterizations and the model-based approaches used to solve them. We then propose OpinionRank, a model-free, interpretable, graph-based spectral algorithm for integrating crowdsourced annotations into reliable labels for performing supervised or semi-supervised learning. Our experiments show that OpinionRank performs favorably when compared against more highly parameterized algorithms. We also show that OpinionRank is scalable to very large datasets and numbers of label sources, and requires considerably fewer computational resources than previous approaches.
Comments: 8 pages, 5 figures, accepted at IJCNN 2021
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Cite as: arXiv:2102.05884 [cs.LG]
  (or arXiv:2102.05884v2 [cs.LG] for this version)

Submission history

From: Glenn Dawson [view email]
[v1] Thu, 11 Feb 2021 08:12:44 GMT (347kb,D)
[v2] Sat, 15 May 2021 03:59:53 GMT (346kb,D)

Link back to: arXiv, form interface, contact.