Time Series Mining through Feature Extraction and Motif Identification

• André Gustavo Maletzke USP / UNIOESTE / PTI
• Gustavo E. A. P. A. Batista USP
• Huei Diana Lee UNIOESTE / PTI
• Feng Chung Wu UNIOESTE / PTI

Abstract

One of the most important challenges in machine learning is the integration of temporal data to the data mining process. However, there is a lack of methods able to induce symbolic and intelligible knowledge from this data, resulting that time series data are usually treated in an adhoc manner. This work proposes a methodology to extract knowledge from time series data using characteristic extraction and motif identification. This methodology aims to induce more precise models and more comprehensible symbolic models, in particular. In addition, the methodology uses a less computationally intensive method to search for motifs. The experimental results obtained were significantly better when compared with one of the most used approaches to classify time series.

References

Chiu, B., Keogh, E., and Lonardi, S. (2003). “Probabilistic discovery of time series motifs”. In: 9th International Conference on Knowledge Discovery and Data Mining, pages 493–498, New York, USA. ACM Press.

Kadous, M. W. and Sammut, C. (2004). “Constructive induction for classifying time series”. In: 15th European Conference on Machine Learning, volume 3201 of Lecture Notes in Computer Science, pages 192–204, Pisa, Italy. Springer.

Last, M., Klein, Y., and Kandel, A. (2001). “Knowledge discovery in time series databases”. IEEE Transactions on Systems, Man, and Cybernetics, Part B, 31(1):160–169.

Lin, J., Keogh, E., Lonardi, S., and Chiu, B. (2003). “A symbolic representation of time series, with implications for streaming algorithms”. In: 8th Workshop on Research Issues in Data Mining and Knowledge Discovery, pages 2–11, San Diego, USA. ACM Press.

Maletzke, A. G., Batista, G. E., and Lee, H. D. (2008). “Uma avaliação sobre a identificação de motifs em séries temporais”. In: 3th Congresso da Academia Trinacional de Ciências, volume 1, pages 1–10, Foz do Iguaçu, Brasil.

Michalski, R. S., Bratko, I., and Kubat, M. (1998). Machine learning and data mining. Wiley, Chichester, West Sussex, England.

Weiss, S. M. and Indurkhya, N. (1998). Predictive Data Mining: a practical guide. Morgan Kaufmann, California, USA.

Yang, Q. and Wu, X. (2006). “10 challenging problems in data mining research”. International Journal of information Technology and Decision Making, 5(4):401–420.

Yankov, D., Keogh, E., Medina, J., Chiu, B., and Zordan, V. (2007). “Detecting time series motifs under uniform scaling”. In: 13th International Conference on Knowledge Discovery and Data Mining, pages 844–853, New York, USA. ACM Press.