Imbalanced Multiple Noisy Labeling for supervised learning

Jing Zhang, Xindong Wu, Victor S. Sheng

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

When labeling objects via Internet-based outsourcing systems, the labelers may have bias, because they lack expertise, dedication and personal preference. These reasons cause Imbalanced Multiple Noisy Labeling. To deal with the imbalance labeling issue, we propose an agnostic algorithm PLAT (Positive LAbel frequency Threshold) which does not need any information about quality of labelers and underlying class distribution. Simulations on eight realworld datasets with different underlying class distributions demonstrate that PLAT not only effectively deals with the imbalanced multiple noisy labeling problem that off-theshelf agnostic methods cannot cope with, but also performs nearly the same as majority voting under the circumstances that labelers have no bias.

Original languageEnglish
Title of host publicationProceedings of the 27th AAAI Conference on Artificial Intelligence, AAAI 2013
Pages1651-1652
Number of pages2
StatePublished - 2013
Event27th AAAI Conference on Artificial Intelligence, AAAI 2013 - Bellevue, WA, United States
Duration: Jul 14 2013Jul 18 2013

Publication series

NameProceedings of the 27th AAAI Conference on Artificial Intelligence, AAAI 2013

Conference

Conference27th AAAI Conference on Artificial Intelligence, AAAI 2013
CountryUnited States
CityBellevue, WA
Period07/14/1307/18/13

Fingerprint Dive into the research topics of 'Imbalanced Multiple Noisy Labeling for supervised learning'. Together they form a unique fingerprint.

Cite this