Effect of Label Noise on the Machine-Learned Classification of Earthquake Damage

Effect of Label Noise on the Machine-Learned Classification of Earthquake Damage

Automated classification of earthquake damage in remotely-sensed imagery using machine learning techniques depends on training data, or data examples that are labeled correctly by a human expert as containing damage or not. Mislabeled training data are a major source of classifier error due to the u...

Full description

Bibliographic Details
Main Authors: Jared Frank, Umaa Rebbapragada, James Bialas, Thomas Oommen, Timothy C. Havens
Format: Article
Language:English
Published: MDPI AG 2017-08-01
Series:Remote Sensing
Subjects:
machine learning
classification
crowdsourcing
earthquake damage
damage detection
GEOBIA
mislabeled training data
Online Access:https://www.mdpi.com/2072-4292/9/8/803

Internet

https://www.mdpi.com/2072-4292/9/8/803

Similar Items