No Access

NEW SELF-ORGANIZING MAPS FOR MULTIVARIATE SEQUENCES PROCESSING

FARIDA ZEHRAOUI

LIPN-CNRS, Université Paris 13 — Institut Galilée, 99, Avenue J-B. Clément — 93430 Villetaneuse, France

Search for more papers by this author

and

YOUNÉS BENNANI

LIPN-CNRS, Université Paris 13 — Institut Galilée, 99, Avenue J-B. Clément — 93430 Villetaneuse, France

Search for more papers by this author

https://doi.org/10.1142/S1469026805001684Cited by:3 (Source: Crossref)

Abstract

Spatio-temporal connectionist networks comprise an important class of neural models that can deal with patterns distributed in both time and space. In this article, we present new models of self-organizing maps for sequence clustering and classification. We have introduced the temporal dynamics in these maps and we have proposed several new models based on covariance matrices computation. In the first models, the inputs are modeled using its associated covariance matrix. These models, used in speaker recognition, do not take into account the order of the vectors in the sequence. To overcome this drawback, we have proposed new models, which introduce the temporal dynamics in the covariance matrix associated to the input sequences. In order to obtain a network that can learn new knowledge without forgetting the previous learned ones, we have introduced the plasticity and stability properties into one proposed temporal model using the adaptive resonance theory paradigm.

Keywords:

Remember to check out the Most Cited Articles!
Check out these titles in artificial intelligence!

References

A. Baraldi and E. Alpaydin, Simplified ART: A new class of ART algorithms, Technical Report TR-98-004, Berkeley, CA (1998) . Google Scholar
G. A. Barreto and A. F. R. Arajo, Int. J. Comput. Res. 10(2), 139 (2001). Google Scholar
Y. Bennani, F. Fogelman Soulie and P. Gallinari, A connexionist approach for automatic speaker identification, Int. Conf. Acoustics Speech Signal Process. "ICASSP" (1990) pp. 265–272, DOI: 10.1109/ICASSP.1990.115619. Google Scholar
F. Bimbot and L. Mathan, Text-free speaker recognition using an arithmetic harmonic sphericity measure, Eurospeech'931 (1993) pp. 169–172. Google Scholar
G. A. Carpenter, Neural Networks 10(8), 1473 (1997), DOI: 10.1016/S0893-6080(97)00004-X. Crossref, Google Scholar
G. A. Carpenter and S. Grossberg, Comput. Vision Graphics Image Process. 37(1), 54 (1987), DOI: 10.1016/S0734-189X(87)80014-2. Crossref, Google Scholar
G. A. Carpenter and S. Grossberg, IEEE Comput. 21(3), 77 (1988). Crossref, Google Scholar
G. A. Carpenter, S. Grossberg and D. B. Rosen, Neural Networks 4, 759 (1991), DOI: 10.1016/0893-6080(91)90056-B. Crossref, Google Scholar
G. A. Carpenter, S. Grossberg and D. B. Rosen, Neural Networks 4, 759 (1991), DOI: 10.1016/0893-6080(91)90056-B. Crossref, Google Scholar
G. J. Chappell and J. G. Taylor, Neural Networks 6, 441 (1993), DOI: 10.1016/0893-6080(93)90011-K. Crossref, Google Scholar
S. Durand and F. Alexandre , Spatio-temporal mask learning: Application to speech recognition , Proc. Int. Conf. Artif. Neural Networks Genetic Algorithms ( 1995 ) . Google Scholar
N. R. Euliano and J. C. Principe, Spatio-temporal self-organising feature maps, Proc. ICNN '961 (1996) pp. 1900–1905. Google Scholar
B. Fritzke, Growing cell structures — A self-organizing network for unsupervised and supervised learning, TR-93-026, International Computer Science Institute, Berkeley (1993) . Google Scholar
B. Fritzke, Neural Process. Lett. 2(5), 9 (1995), DOI: 10.1007/BF02332159. Crossref, Google Scholar
B. Fritzke, Adv. Neural Inf. Process. Syst. 7, eds. G. Tesauro, D. S. Touretzky and T. K. Leen (Cambridge, MA, 1995) pp. 625–632. Google Scholar
S. Grossberg, Cognitive Sci. 11(1), 23 (1987), DOI: 10.1016/S0364-0213(87)80025-3. Crossref, Google Scholar
M. Hagenbuchner, A. Sperduti and A. C. Tsoi, IEEE Trans. Neural Networks 14, 491 (2003), DOI: 10.1109/TNN.2003.810735. Crossref, Google Scholar
M. Hagiwara, Neurocomputing 6, 513 (1994), DOI: 10.1016/0925-2312(94)90003-5. Crossref, Google Scholar
D. L. James and R. Miikkulainen, Adv. Neural Process. Syst. 7, eds. D. S. Tourestzky, G. Tesauro and T. K. Leen (MIT Press, Cambridge, MA, USA, 1995) pp. 577–584. Google Scholar
J. Kangas (ed.), Phoneme recognition using time-dependent versions of self-organizing maps, IEEE Int. Conf. Acoustics Speech Signal Process (ICASSP'91)1 (1991) pp. 101–104, DOI: 10.1109/ICASSP.1991.150288. Google Scholar
T. Kohonen, Artif. Neural Network 1, eds. O. Simulaet al. (North-Holland, Amstdam, Netherlands, 1991) pp. 1357–1360. Crossref, Google Scholar
T. Kohonen , Self-Organizing Maps , Springer Series in Information Sciences 30 ( Springer , Berlin, Heidelberg , 1995 ) . Crossref, Google Scholar
J. Lampinen and E. Oja, Self-organizing maps for spatial and temporal AR models, Proc. 6 SCIA, Scand. Conf. Image Anal., eds. M. Pietikäinen and J. Röning (Suomen Hahmontunnistustutkimuksen seura r.y, 1989) pp. 120–127. Google Scholar
W. M. Leung, Y. M. Cheung and L. Xu, Fuzzy arttsp: Application of adaptive resonance theory on non-stationary financial time series prediction, Int. Conf. Neural Inf. Process. (ICONIP'1995)1 (1995) pp. 449–452. Google Scholar
F. d'Alché-Buc, M.-J. Lesot and G. Siolas, Evaluation of topographic clustering and its kernelization, Proc. Eur. Conf. Mach. Learn. ECML03 (Springer, Dubrovnik, Croatie, 2003) pp. 265–276. Google Scholar
C. M. Privitera and L. Shastri, Temporal compositional processing by a dsom hierarchical model, Int. Conf. Artif. Neural Network (ICANN'96) (1996) pp. 457–462. Google Scholar
P. Somervuo, Competing hidden Markov models on the self-organizing map, Int. Joint Conf. Neural Networks3 (Helsinki University of Technology, Piscataway, NJ, 2000) pp. 169–174. Google Scholar
P. Somervuo and T. Kohonen, Neural Process. Lett. 10(2), 151 (1999), DOI: 10.1023/A:1018741720065. Crossref, Google Scholar
M. Strickert and B. Hammer, Self-organizing context learning, Eur. Symp. Artif. Neural Networks' 2004, ed. M. Verleysen (2004) pp. 39–44. Google Scholar
T. Voegtlin, Neural Networks 15(8–9), 979 (2002), DOI: 10.1016/S0893-6080(02)00072-2. Crossref, Google Scholar
A. Waibelet al., IEEE Trans. Acoustics Speech Signal Process. 37(3), (1989). Google Scholar
J. C. Wiemer, Learning Topography in Neural Networks: Towards a Better Understanding of Cortical Topography, PhD thesis, Shaker Verlag (2001) . Google Scholar
J. C. Wiemer, Neural Comput. 15(5), 1143 (2003), DOI: 10.1162/089976603765202695. Crossref, Google Scholar
A. Zeboulon , Y. Bennani and K. Benabdeslem , Hybrid connectionist approach for knowledge discovery from web navigation patterns , Proc. ACS/IEEE Int. Conf. Comput. Syst. Appl. ( Tunisia , 2003 ) , DOI: 10.1109/AICCSA.2003.1227550 . Google Scholar
F. Zehraoui, Systmes d'apprentissage connexionnistes et raisonnement partir de cas pour la classification et le classement de squences. PhD thesis, Paris (13 October 2004) . Google Scholar
F. Zehraoui and Y. Bennani, M-som-art: Growing self organizing map for sequences clustering and classification, 16th Eur. Conf. Artif. Intell. (ECAI2004) (2004) pp. 564–570. Google Scholar
F. Zehraoui and Y. Bennani, M-som: Matricial self organizing map for sequences clustering and classification, Int. Joint Conf. Neural Networks (IJCNN 2004) (2004) pp. 763–768. Google Scholar