Abstract
We present a non-overlapping serial Episodes Based Hierarchical Hidden Markov Model (EBHHMM). Each EBHHMM is associated with two types of temporal frequent patterns, i.e. non-overlapping serial episodes and the presented SEI (Serial Episode Interactions). Serial episode interaction is a set of non-overlapping serial episodes which are correlated and occurs frequently in sequence. As the key advantage of our approach, we do not need any prior-knowledge to learn the structure of EBHHMM. Extensive experiments perform on real world data demonstrate that EBHHMM gets larger maximum log likelihood (has better quality) than existing models.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Durbin, R., Eddy, S., Krogh, A., et al.: Biological Sequence Analysis. Cambridge University Press, Cambridge (1998)
Jensen, K., Styczynski, M., Rigoutsos, I., et al.: A generic motif discovery algorithm for sequential data. Bioinform. 22, 21–28 (2006)
Nanopolulos, A., Katsaros, D., Manolopoulos, Y.: A data mining algorithm for generalized Web prefetching. IEEE Trans. Knowl. Data Engin. 15(5), 1155–1169 (2003)
Laxman, S., Sastry, P.S., Unnikrishnan, K.P.: Discovering frequent episodes and learning Hidden Markov Models: A formal connection. IEEE Transactions on Knowledge and Data Engineering 17(11), 1505–1517 (2005)
Laxman, S.: Stream Prediction Using A Generative Model Based On Frequent Episode. In: Event Sequences Proceeding of KDD (2008)
Zaki, M.J., Carothers, C.C., Szymanski, B.K.: VOGUE: A Variable Order Hidden Markov Model with Duration based on Frequent Sequence Mining. ACM Transactions on Knowledge Discovery From Data 4(1) (2010) Article 5
Fine, S., Singer, Y., Tishby, N.: The Hierarchical Hidden Markov Model: Analysis and Applications. Machine Learning 32, 41–62 (1998)
Carl, H.M., John, F.R.: Mining relationships between interacting episodes. SIAM, Philadelphia
Kuramochi, M., Karypis, G.: Finding frequent patterns in a large sparse graph. Data Mining and Knowledge Discovery (2005)
Molecular Biology (Splice-junction Gene Sequences) Data Set, http://archive.ics.uci.edu/ml/datasets/
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wan, L. (2011). Learning Frequent Episodes Based Hierarchical Hidden Markov Models in Sequence Data. In: Shen, G., Huang, X. (eds) Advanced Research on Computer Science and Information Engineering. CSIE 2011. Communications in Computer and Information Science, vol 153. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-21411-0_19
Download citation
DOI: https://doi.org/10.1007/978-3-642-21411-0_19
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-21410-3
Online ISBN: 978-3-642-21411-0
eBook Packages: Computer ScienceComputer Science (R0)