Abstract
For the problem that large-scale labeled samples are not easy to acquire in the course of Support Vector Machines (SVMs) training, a Semi-Supervised Active Learning Algorithm for SVMs (QTB-ASVM) is proposed in the paper, which efficiently combines the semi–supervised learning based on Tri-Training and active learning based on Query By Committee (QBC) with SVMs. With this method, QBC active learning is used to select the samples which are the most valuable to current SVM classifier, and Tri-Training is used to exploit useful information that remains in the unlabeled samples. The experimental results show that the proposed approach can considerably reduce the labeled samples and costs compared to the SVMs which is either not applied with semi-supervised learning or active learning or applied with only one of them, and at the same time it can ensure that the accurate classification performance is kept as the passive SVM, while improving generalization performance and also expediting the SVM training.
Similar content being viewed by others
References
Brefeld , Tobias S (2004) Co-EM support vector learning. In: Proceedings of the 21th International Conference on Machine Learning. ACM New York, NY, USA, pp 16–24
Fang Y, Chunyang H, Zhang LL et al (2016) Breakout prediction classifier for continuous casting based on active learning GA-SVM. Chin Mech Eng 27(12):1609–1914
Gokhan T, Hakkani-Tűr D, Schapire RE (2005) Combining active and semi-supervised learning for spoken language understanding. Speech Commun 45(2):171–186
Goudjil M, Koudil M, Bedda M et al (2018) A novel active learning method using SVM for text classification. Int J Autom Comput 15(3):290–298
Handing W, Yaochu J, John D (2017) Committee-based active learning for surrogate-assisted particle swarm optimization of expensive problems. IEEE Trans Cybern 47(9):2664f–22677
Hoi SCH, Lyu MR (2005) A semi-supervised active learning frame work for image retrieval. In: Proceedings of 2005 IEEE computer society conference on computer vision and pattern recognition, Los Angels, IEEE Computer Society pp 302–309
Ion M, Steven M, Knoblock CA (2002) Active + semi-supervised learning = robust multi-view learning. In: Proceedings of the nineteenth international conference on machine learning, pp 435–442
Jie XU, Pf SHI (2004) Active learning with labeled and unlabeled samples for content-based image retrieval. J Shang Hai Jiao Tong Univ 38(12):2068–2072
Kothari R, Jain V (2003) Learning from labeled and unlabeled data using a minimal number of queries. IEEE Trans Neural Netw 14(6):1496–1505
Li LY, Liu L FXGZ et al (2016) Modified particle swarm optimization for BMDS interceptor resource planning. Appl Intell 44(3):471–488
Li LY, Liu FX, L GZ et al (2016) Performance analysis and optimal allocation of layered defense M/M/N queueing systems. Math Probl Eng pp 1–32
McCallum AK, Nigam K (1998) Employing EM and pool-based active learning for text classification. In: Proceeding of the 15th international conference on machine learning, Morgan Kaufmann Publishers, USA, pp 350–358
Seung HS, Opper M, Sompolinsky H (1992) Query by committee. In: Proceedings of the 15th annual ACM conference on computational learning theory, Pittsburgh, Morgan Kauf mann, pp 287~294
Singla A, Patra S (2018) A fast partition-based batch-mode active learning technique using SVM classifier. Soft Comput 22(14):4627–4637
Tur G, Hakkani-Tűr D, Schapire RE (2005) Combining active and semi-supervised learning for spoken language understanding. Speech Commun 45(2):171–186
Wang XG, Liu LX (2017) Research on building bank anti-fraud model based on tri-training semi supervised learning and fuzzy SVM active learning. In: 2017 3rd Workshop on Advanced Research and Technology in Industry Applications (WARTIA 2017), Atlantis Press, pp 383–387
Wang XJ, Luo GC, Ke Q, Aiguo C et al (2016) Image retrieval method based on SVM and active learning. Appl Res Comput 33(12):3836–3838
Xu HL, Wang XD, Liao Y et al (2010a) New approach for optimizing model of RBF-SVM based on PSO. Control Decis 25(3):367–370
Xu HL, Wang XD, Liao Y et al (2010b) Incremental training algorithm of SVM based on active learning. Control and Decision 25(2):282–286
Xu HL, Bie XF, Feng H et al (2015) Active learning algorithm for SVM based on QBC. Syst Eng Electron 37(12):2865–2871
Xu HL, Long GZ, Bie XF et al (2016) Active learning algorithm of SVM combining tri-training semi-supervised learning and convex-hull vector. Pattern Recognit Artif Intell 29(1):39–46
Xu Z, Cheng C, Sugumaran V (2020) Big data analytics of crime prevention and control based on image processing upon cloud computing. J Surveill Secur Saf 1:16–33
Yan-fei P, Yong-gang S, De-jian W (2014) A novel retrieval method based on SVM and active learning. Comput Eng Sci 36(7):1371–1376
Zhang JM, Sun CM, Yan T (2013) Video Annotation for semi-supervised active learning based on adaptive SVM. Comput Eng 39(8):190–195
Zhao WZ, Ma HF, Li ZQ et al (2012) Efficient active learning for semi-supervised document clustering. J Softw 23(6):1486–1499
Zhou ZH, Li M (2005) Tri-training: exploiting unlabeled data using three classifiers. IEEE Trans Knowl Data Eng 17(11):1529–1541
Acknowledgement
The authors wish to express their gratitude to the referees for their helpful comments and kind suggestions in revising this paper. This work is substantially supported by grants from the Natural Science Foundation of China (Nos. 71701209, 72071209), and the Natural Science Foundation of Shaanxi Province of China (Nos. 2019JQ-250), and the China Postdoctoral Science Foundation funded project (2017M613415, 2019M653962).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Xu, H., Li, L. & Guo, P. Semi-supervised active learning algorithm for SVMs based on QBC and tri-training. J Ambient Intell Human Comput 12, 8809–8822 (2021). https://doi.org/10.1007/s12652-020-02665-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-020-02665-w