Abstract
Lacking labeled samples is an important bottleneck in the development of novelty detection in practical industrial applications. To solve this problem, this paper proposes a novel novelty detection method called active learning-based support vector data description (ALSVDD). Here, we combine the uncertainty information and the importance of each sample to guide the selection process of active learning. In addition, we propose a simple recursive sequential minimal optimization (SMO) strategy to solve the ALSVDD optimization problem. Finally, the experiments carried out on the UCI data sets prove the effectiveness of the proposed method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abe N, Zadrozny B, Langford J (2006) Outlier detection by active learning. In: Proceedings of the 12th ACM SIGKDD international conference on knowledge discovery and data mining. Philadelphia, PA, USA
Almgren M, Jonsson E (2004) Using active learning in intrusion detection. In: Proceedings 17th IEEE Computer Security Foundations Workshop
Blake C, Merz CJ (1998) UCI repository of machine learning databases. http://www.ics.uci.edu/~mlearn/MLRepository.html
Chen G, Zhang X, Wang ZJ, Li F (2015) Robust support vector data description for outlier detection with noise or uncertain data. Know Based Sys. doi:10.1016/j.knosys.2015.09.025
Fan R, Chen P, Lin C (2005) Working set selection using second order information for training support vector machines. J mach learn res 6(12):1889–1918
Faria ER, Gonçalves IJ, de Carvalho AC, Gama J (2016) Novelty detection in data streams. Artif Intell Rev. doi:10.1007/s10462-015-9444-8
Görnitz N, Kloft M, Brefeld U (2009) Active and semi-supervised data domain description. Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Bled, Slovenia
Li D, Cai J, Du M, Zhu S, Zhang J (2015) SVDD fast training algorithm based on improved SMO. China Meas Text. doi:10.11857/j.issn.1674-5124.2015.11.022
Li Y, Guo L (2007) An active learning based TCM-KNN algorithm for supervised network intrusion detection. Comp Secur. doi:10.1016/j.cose.2007.10.002
Liu Y, Liu Y, Chen Y (2010) Fast support vector data descriptions for novelty detection. IEEE Trans Neural Networks. doi:10.1109/TNN.2010.2053853
Pelleg D, Moore AW (2004) Active learning for anomaly and rare-category detection. NIPS, Vancouver Canada
Schölkopf B, Burges CJ, Smola AJ (eds) (1999) Advances in kernel methods: support vector learning. MIT press
Seliya N, Khoshgoftaar TM (2010) Active learning with neural networks for intrusion detection. IEEE International Conference on Information reuse and integration (IRI). Las Vegas, Nevada
Xiao Y, Wang H, Zhang L, Xu W (2014) Two methods of selecting Gaussian kernel parameters for one-class SVM and their application to fault detection. Knowl Based Sys. doi:10.1016/j.knosys.2014.01.020
Xie Y, Chen X, Yu X, Yue B, Guo J (2011) Fast SVDD-based outlier detection approach in wireless sensor networks. Chinese J Sci Instr 1:009. (in Chinese)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Yin, L., Wang, H., Fan, W., Wang, Q. (2018). Active Learning Based Support Vector Data Description for Large Data Set Novelty Detection. In: Deng, Z. (eds) Proceedings of 2017 Chinese Intelligent Automation Conference. CIAC 2017. Lecture Notes in Electrical Engineering, vol 458. Springer, Singapore. https://doi.org/10.1007/978-981-10-6445-6_32
Download citation
DOI: https://doi.org/10.1007/978-981-10-6445-6_32
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-6444-9
Online ISBN: 978-981-10-6445-6
eBook Packages: EngineeringEngineering (R0)