|
|
|
| A Batch-mode Active Learning Method Based on the Nearest Average-class Distance (NACD) for Multiclass Brain-Computer Interfaces
|
| State Key Laboratory of Power Transmission Equipment & System Security and New Technology
School of Electrical Engineering, Chongqing University, Chongqing 400044, China
School of Computer Science and Electronic Engineering, University of Essex Colchester, CO4 3SQ, UK
|
|
|
|
|
Abstract In this paper, a novel batch-mode active learning method based on the nearest average-class distance
(ALNACD) is proposed to solve multi-class problems with Linear Discriminate Analysis (LDA) classifiers.
Using the Nearest Average-class Distance (NACD) query function, the ALNACD algorithm selects
a batch of most uncertain samples from unlabeled data to improve gradually pre-trained classifiers'
performance. As our method only needs a small set of labeled samples to train initial classifiers, it is
very useful in applications like Brain-computer Interface (BCI) design. To verify the effectiveness of the
proposed ALNACD method, we test the ALNACD algorithm on the Dataset 2a of BCI Competition IV.
The test results show that the ALNACD algorithm offers similar classification results using less sample
labeling effort than Random Sampling (RS) method. It also provides competitive results compared with
active Support Vector Machine (active SVM), but uses less time than the active SVM in terms of the
training.
|
|
|
| Fund:Project supported by the Fundamental Research Funds for the Central Universities in China (Project No. CD-
JZR13150010) and National \111" Plan Project (B08036).
|
|
|
|
| Cite this article: |
|
Minyou Chen,Xuemin Tan,John Q.Gan, et al. A Batch-mode Active Learning Method Based on the Nearest Average-class Distance (NACD) for Multiclass Brain-Computer Interfaces
[J]. Journal of Fiber Bioengineering and Informatics, 2014, 7(4): 627-636.
|
|
| [1] Ang KK, Chin ZY, Wang C, Guan C and Zhang H. Filter bank common spatial pattern algorithm
on BCI competition iv datasets 2a and 2b. Front Neurosci: 2012; 6-39.
[2] Liu C, Wang H, Lu Z. EEG classi¯cation for multi-class motor imagery BCI. In Proc. of the 25th
Chinese Control and Decision Conference (CCDC): 2013; 4450-4453.
[3] Nguyen TT, Binh ND and Bischof H. E±cient boosting-based active learning for specific object
detection problems. International Journal of Electrical, Computer and Systems Engineering: 2009;
3: 150-155.
636 M. Chen et al. / Journal of Fiber Bioengineering and Informatics 7:4 (2014) 627{636
[4] Yang T, Li J, Pan Q, Zhao C and Zhu Y. Active learning based pedestrian detection in real scenes.
In Proc. of the 18th International Conference on Pattern Recognition: 2006; 904-907.
[5] Mitra P, Uma Shankar B, Pal K. Segmentation of multispectral remote sensing images using active
support vector machines. Pattern Recogn Lett: 2004; 25: 1067-1074.
[6] Pasolli E, Melgani F. Active learning methods for electrocardiographic signal classification. IEEE
T Inf Technol B: 2010; 14: 1405-1416.
[7] Cai D, He X and Han J. Semi-supervised discriminant analysis. In Proc. of the IEEE 11th Inter-
national Conference on Computer Vision: 2007; 1-7.
[8] Zhao X, Evans N and Dugelay J. Semi-supervised face recognition with LDA self-training. In Proc.
of the 18th IEEE International Conference on Image Processing: 2011, 3041-3044.
[9] Zheng Z and Peng Y. Semi-supervised based hyperspectral imagery classification. Sensors & Trans-
ducers: 2013; 156: 298-303.
[10] Tong S and Koller D. Support vector machine active learning with applications to text classifica-
tion. J Mach Learn Res: 2002; 2: 45-66.
[11] Lewis DD and Gale WA. A sequential algorithm for training text classifiers. In Proc. of the
17th Annual International ACM SIGIR Conference on Research and Development in Information
Retrieval: 1994; 3-12.
[12] Guo Y. Active instance sampling via matrix partition. NIPS: 2010; 802-810.
[13] Guo Y, Schuurmans D. Discriminative Batch Mode Active Learning. NIPS: 2007.
[14] Xanthopoulos P, Pardalos PP and Trafalis TB. Robust Data Mining. Springer: 2013.
[15] Naeem M, Brunner C, Leeb R, Graimann B and Pfurtscheller G., Seperability of four-class motor
imagery data using independent components analysis. J Neural Eng: 2006; 3: 208.
[16] Wang H, Tang Q and Zheng W. L1-norm-based common spatial patterns. IEEE T Biomedical
Enginee: 2012; 59: pp. 653-662.
[17] Fanga Y, Chena M, Zhengc X and Harrisond RF. Extending CSP to Detect Motor Imagery in a
Four-class BCI. Journal of Information & Computational Science: 2012; 9: 143-151.
[18] Muller-Gerking J, Pfurtscheller G, and Flyvbjerg H. Designing optimal spatial ¯lters for single-trial
EEG classi¯cation in a movement task. Clin. Neurophysiol: 1999; 110: 787-798.
[19] Chang CC and Lin CJ. LIBSVM: a library for support vector machines. ACM Transactions on
Intelligent Systems and Technology (TIST): 2011; 2: pp. 27.
[20] Xiong T and Cherkassky V. A combined SVM and LDA approach for classification. In Proc. of
the IEEE International Joint Conference on Neural Networks: 2005; 1455-1459.
|
|
|
|
2014, Vol. 7 
