Year: 2014
Type: conference-paper
Abstract: Active learning aims to label the most informative data points in order to minimize \nthe cost of labeling [1]. In this work, we introduce a human based approach, namely First Certain Wrong Labeled (FCWL) to select points for labeling. It is based on a ranked list of predictions ordered by confidence, from which the user selects the highest ranking incorrect prediction. The experimental results show the improvement in performance of this method compared to others.
Source: elib (German Aerospace Center)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2193997355