Abstract
In this paper, a novel semi-supervised feature extraction algorithm, i.e., semi-supervised transfer discriminant analysis (STDA) with knowledge transfer capability is proposed, based on the traditional algorithm that cannot get adapted in the change of the learning environment. By using both the pseudo label information from target domain samples and the actual label information from source domain samples in the label iterative refinement process, not only the between-class scatter is maximized while that within-class scatter is minimized, but also the original space structure is maintained via Laplacian matrix, and the distribution difference is reduced by using maximum mean discrepancy as well. Moreover, semi-supervised transfer discriminant analysis based on cross-domain mean constraint (STDA-CMC) is proposed. In this algorithm, the cross-domain mean constraint term is incorporated into STDA, such that knowledge transfer between domains is facilitated by making source and target samples after being projected are located more closely in the low-dimensional feature subspace. The proposed algorithm is proved efficient and feasible from experiments on several datasets.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bharti KK, Singh PK (2013) A two-stage unsupervised dimension reduction method for text clustering. In: BIC-TA, pp 529–542
Borgwardt KM, Gretton A, Rasch MJ, Kriegel HP, Schölkopf B, Smola AJ (2006) Integrating structured biological data by kernel maximum mean discrepancy. Bioinformatics 22(14):E49–E57
Cai D, He XF, Han JW (2007) Semi-supervised discriminant analysis. In: ICCV, pp 1–7
Cai D, He XF, Han JW (2011) Speed up kernel discriminant analysis. VLDB J 20(1):21–33
Chen J, Liu Y (2011) Locally linear embedding: a survey. Artif Intell Rev 36(1):29–48
Gong BQ, Shi Y, Sha F, Grauman K (2012) Geodesic flow kernel for unsupervised domain adaptation. In: CVPR, pp 2066–2073
Har-Peled S, Indyk P, Motwani R (2012) Approximate nearest neighbor: towards removing the curse of dimensionality. Theory Comput 8(1):321–350
He XF, Niyogi P (2003) Locality preserving projections. In: NIPS, pp 153–160
He XF, Cai D, Yan SC, Zhang HJ (2005) Neighborhood preserving embedding. In: ICCV, pp 1208–1213
Liu A, Chen MY, Zhou DH (2015) Discriminative dictionary learning with low-rank error model for robust crater recognition. IEICE Trans Inf Syst 98(5):1116–1119
Long MS, Wang JM, Ding GG, Sun JG, Yu, PS (2013) Transfer feature learning with joint distribution adaptation. In: ICCV, pp 2200–2207
Lv L, Zhao DB, Deng QQ (2016) A semi-supervised predictive sparse decomposition based on the task-driven dictionary learning. Cognitive Comput. doi:10.1007/s12559-016-9438-0
Pan SJ, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359
Pan SJ, Tsang IW, Kwok JT, Yang Q (2011) Domain adaptation via transfer component analysis. IEEE Trans Neural Netw 22(2):199–210
Shi L, Zhang LF, Zhao LL, Zhang LP, Li PX, Wu D (2016) Adaptive laplacian eigenmap-based dimension reduction for ocean target discrimination. IEEE Trans Geo Remot Sens Lett 13(7):902–906
Shu L, Latecki LJ (2015) Transductive domain adaptation with affinity learning. In: CIKM, pp 1903–1906
Si S, Tao DC, Geng B (2010) Bregman divergence-based regularization for transfer subspace learning. IEEE Trans Knowl Data Eng 22(7):929–942
Tong B, Jia WF, Ji YL, Suzuki E (2012) Linear semi-supervised dimensionality reduction with pairwise constraint for multiple subclasses. IEICE Trans Inf Syst E95D:812–820
Wan MH, Lai ZH, Jin Z (2011) Feature extraction using two-dimensional local graph embedding based on maximum margin criterion. Appl Math Comput 217(23):9659–9668
Xing XL, Du SD, Jiang H (2013) Semi-supervised nonparametric discriminant analysis. IEICE Trans Inf Syst 96(2):375–378
Yamamoto S, Hosoya Y, Tsumura N, Ogawa-Ochiai K (2012) Principal component analysis for dental shade color. Den Mater 28(7):736–742
Zhang H, Yang ZR, Oja E (2012) Adaptive multiplicative updates for projective nonnegative matrix factorization. In: ICONIP, pp 277–284
Zhou W, Kamata SI (2013) L1-norm based linear discriminant analysis: an application to face recognition. IEICE Trans Inf Syst E96–D(3):550–558
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the National Natural Science Foundation of China under Grant 61273143, by the Fundamental Research Funds for the Central Universities under Grant 2013RC12.
Rights and permissions
About this article
Cite this article
Zang, S., Cheng, Y., Wang, X. et al. Semi-supervised transfer discriminant analysis based on cross-domain mean constraint. Artif Intell Rev 49, 581–595 (2018). https://doi.org/10.1007/s10462-016-9533-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10462-016-9533-3