A hybrid approach to dimension reduction in classification

• Autorzy: Krawczak M. , Szkatuła G.
• Języki publikacji: EN

Abstrakty

EN

In this paper we introduce a hybrid approach to data series classification. The approach is based on the concept of aggregated upper and lower envelopes, and the principal components here called 'essential attributes', generated by multilayer neural networks. The essential attributes are represented by outputs of hidden layer neurons. Next, the real valued essential attributes are nominalized and symbolic data series representation is obtained. The symbolic representation is used to generate decision rules in the IF. . . THEN. . . form for data series classification. The approach reduces the dimension of data series. The efficiency of the approach was verified by considering numerical examples.

Słowa kluczowe

EN

data series; dimension reduction; envelopes; essential attributes; heteroassociation; machine learning from examples; decision rules; classification

• Wydawca: Systems Research Institute, Polish Academy of Sciences
• Czasopismo: Control and Cybernetics
• Rocznik: 2011
• Tom: Vol. 40, no 2
• Strony: 527--551
• Opis fizyczny: Bibliogr. 28 poz.

Twórcy

autor

Krawczak M.

autor

Szkatuła G.

• Systems Research Institute, Polish Academy of Sciences ul. Newelska 6, 02-776 Warsaw, Poland,

Bibliografia

• Alcock,R.J. and Manolopoulos,Y. (1999) Time-Series Similarity Queries Employing a Feature-Based Approach. 7th Hellenic Conference on Informatics, Ioannina, Greece.
• Benedikt, L., Kajic, V., Cosker, D., Marshall, D. and Rosin, P.L. (2008) Facial Dynamics in Biometric Identification. Proc. of British Machine Vision Conference, Leeds, 2008.
• Chan K. and Fu A.W. (2005) Efficient time series matching by wavelets. Proceedings 15th IEEE International Conference on Data Engineering. Sydney, Australia, 126-133.
• Choy,E.,Krawczak,M., Shannon,A. and Szmidt,E., eds. (2007) A Survey of Generalized Nets. KvB Institute of Technology, Sydney, Australia.
• Faloutsos,C., Ranganathan,M. and Manolopulos,Y. (1994) Fast subsequence matching in time-series databases. SIGMOD Record, 23, 519-529.
• Jolliffe, I.T. (2002) Principal Component Analysis. Springer.
• Kacprzyk, J. and Szkatuła,G. (1999) An inductive learning algorithm with a preanalysis data. International Journal of Knowledge - Based Intelligent Engineering Systems, 3, 135-146.
• Kacprzyk, J. and Szkatuła, G. (2002) An integer programming approach to inductive learning using genetic and greedy algorithms. In: L.C. Jain and J.Kacprzyk, eds., New Learning Paradigms in Soft Computing. Studies in Fuzziness and Soft Computing. Physica-Verlag Heidelberg, 323-367.
• Kacprzyk, J. and Szkatuła, G. (2005a) A softened formulation of inductive learning and its use for coronary disease data. Lecture Notes in Artificial Intelligence, 3488, 200-209.
• Kacprzyk, J. and Szkatuła, G. (2005b) An inductive learning algorithm with a partial completeness and consistence via a modified set covering problem. Lecture Notes in Computer Science, 3697, 661-666.
• Kacprzyk, J. and Szkatuła, G. (2010) Inductive Learning: A Combinatorial Optimization. In: J. Koronacki, Z.W. Ras, S.T. Wierzchoń, J. Kacprzyk, eds., Advances in Machine Learning. Studies in Computational Intelligence, 262, Springer.
• Keogh,E., Chakrabarti,K. and Pazzani,M. (2001) Locally Adaptive Dimensionality Reduction for Indexing Large Time Series Databases. In: Proc. of ACM SIGMOD Conference on Management of Data. Santa Barbara. May 21-24, 151-162.
• Krawczak, M. (2003a) Multilayer Neural Systems and Generalized Net Models. Ac. Publ. House EXIT, Warsaw.
• Krawczak, M. (2003b) Heuristic dynamic programming - Learning as control problem. In: L. Rutkowski, J. Kacprzyk, eds., Neural Networks and Soft Computing. Physica Verlag, Heidelberg, 218-223.
• Krawczak, M. (2006) A novel modelling methodology: generalized nets. In: A Cader, L. Rutkowski, R. Tadeusiewicz, J. Żurada, eds., Artificial Intelligence and Soft Computing. Ac. Publ. House EXIT, Warsaw.
• Krawczak, M. and Szkatuła, G. (2008) On decision rules application to time series classification. In: K.T. Atanassov et al., eds., Advances in Fuzzy Sets, Intuitionistic Fuzzy Sets, Generalized Nets and Related Topics, Ac. Publ. House EXIT.
• Krawczak, M., Szkatuła, G. (2010a) Time series envelopes for classification. IEEE Intelligent Systems Conference, London, July 7-9, 2010.
• Krawczak, M. and Szkatuła, G. (2010b) On time series envelopes for classification problems. In: K.T. Atanassov et al., eds., Developments in Fuzzy Sets, Intuitionistic Fuzzy Sets, Generalized Nets and Related Topics. SRI PAS, Warsaw.
• Krawczak, M. and Szkatuła, G. (2010c) On time series clustering. Technical Report. Systems Research Institute PAS, Warsaw.
• Lin, J., Keogh, E., Patel, P. and Lonardi, S. (2002) Finding motifs in time series. The 2nd Workshop on Temporal Data Mining, the 8th ACM International Conference on Knowledge Discovery and Data Mining. Edmonton, Canada, 53-68.
• Nanopoulos,A.,Alcock,R. andManolopoulos,Y. (2001) Feature-based Classification of Time-series Data. International Journal of Computer Research, 49-61.
• Rodríguez, J.J. andAlonso,C.J. (2004) Interval and dynamic time warpingbased decision trees. Proceedings of the 2004 ACM symposium on applied computing (SAC), 548-552.
• Shahabi,C.,Tian,X. and Zhao,W. (2000) TSA-tree: A wavelet-based approach to improve the efficiency of multi-level surprise and trend queries. Proceedings of the 12th International Conference on Scientific and Statistical Database Management. Berlin, 55-68.
• Szkatuła, G. (1995) Machine learning from examples under errors in data, Ph.D. Thesis, SRI PAS Warsaw, Poland.
• Szkatuła, G. (2002) Application of modified covering problem in machine learning. In: J. Gutenbaum, ed., Automatics Control Management. SRI PAS, Warsaw, 431-445.
• Szkatuła,G. and Kacprzyk, J. (2005) An inductive learning algorithm with a partial completeness and consistency. In: M. Dramiński, P. Grzegorzewski, T. Trojanowski, S. Zadrożny, eds., Issues in intelligent systems. Models and techniques. EXIT, Warszawa, 229-246.
• Wu, Y. and Chang, E.Y. (2004) Distance-function design and fusion for sequence data. CIKM ‘04, 324-333.
• Xi, X., Keogh, E.J., Shelton, C.R. and Wei, L. and Ratanamahatana, C.A. (2006) Fast time series classification using numerosity reduction. ICML, 2006.