Asymmetric semi-supervised boosting for SVM active learning in CBIR

Authors:
Jun Wu;Zheng-Kui Lin;Ming-Yu Lu
Affiliations:
Dalian Maritime University, Dalian, China;Dalian Maritime University, Dalian, China;Dalian Maritime University, Dalian, China
Venue:
Proceedings of the ACM International Conference on Image and Video Retrieval
Year:
2010

Citing 6
Cited 1

Support vector machine active learning for image retrieval

MULTIMEDIA '01 Proceedings of the ninth ACM international conference on Multimedia
Asymmetric Bagging and Random Subspace for Support Vector Machines-Based Relevance Feedback in Image Retrieval

IEEE Transactions on Pattern Analysis and Machine Intelligence
A survey of content-based image retrieval with high-level semantics

Pattern Recognition
SemiBoost: Boosting for Semi-Supervised Learning

IEEE Transactions on Pattern Analysis and Machine Intelligence
Semi-Supervised Learning

Semi-Supervised Learning
Bootstrapping SVM active learning by incorporating unlabelled images for image retrieval

CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition

Active SVM-based relevance feedback using multiple classifiers ensemble and features reweighting

Engineering Applications of Artificial Intelligence

Quantified Score

Hi-index	0.01

Visualization

Abstract

Support vector machine (SVM) based active learning technique has played a key role to alleviate the burden of labeling in relevance feedback. However, most SVM-based active learning algorithms are challenged by the small example problem and the asymmetric distribution problem. This paper proposes a novel scheme that combines semi-supervised learning, ensemble learning and active learning in a uniform framework. Concretely, unlabeled data is exploited to facilitate ensemble learning by helping augment the diversity among the base SVM classifiers, and then the learned SVM ensemble model is used to identify the most informative examples for active learning. In particular, a bias-weighting mechanism is developed to guide the ensemble model to pay more attention on the positive examples than the negative ones. An empirical study shows that the proposed scheme is significantly more effective than some existing approaches.