An adaptive streaming active learning strategy based on instance weighting

Mohamed-Rafik Bouguelia

doi:10.1016/j.patrec.2015.11.010

An adaptive streaming active learning strategy based on instance weighting

2016-01-01

28

PRE

AI

M

Mohamed-Rafik Bouguelia *

DOI：10.1016/j.patrec.2015.11.010

Original

Original request for help

Save

Abstract

En 中文

This paper addresses stream-based active learning for classification. We propose a new query strategy based on instance weighting that improves the performance of the active learner compared to the commonly used uncertainty strategies. The proposed strategy computes the smallest weight that should be associated with new instance, so that the classifier changes its prediction regarding this instance. If a small weight is sufficient to change the predicted label, then the classifier was uncertain about its prediction, and the true label is queried from a labeller. In order to determine whether the sufficient weight is small enough, we propose an adaptive uncertainty threshold which is suitable for the streaming setting. The proposed adaptive threshold allows the stream-based active learner to achieve an accuracy which is similar to that of a fully supervised learner, while querying much less labels. Experiments on several public and real world data prove the effectiveness of the proposed method. (C) 2015 Elsevier B.V. All rights reserved.

Keywords:

Classification

Data stream

Active learning

Instance weighting

Uncertainty

An adaptive streaming active learning strategy based on instance weighting

Abstract

AI Summary

Journal

Organization

Cited Papers