ABSTRACT
Emerging patterns are patterns whose frequencies increase from one dataset to another. They can reveal useful trends and contrasts in datasets to support decision making such as trend prediction and classification. However, current works mostly focus on discovering emerging patterns for classification and seldom discuss their use in time-stamped datasets for trend prediction. Though some recent works showed using naive techniques the potential use of emerging patterns in trend prediction, their trend prediction techniques ignore the possible noise or data fluctuations during trend prediction. Additionally, such naive techniques are only able to predict the continuous emergence of patterns but not their supports with time. To effectively use emerging and decaying patterns for trend prediction, we propose EDTrend, a methodology for trend prediction in time-stamped datasets based on emerging and decaying patterns. We show in real-world datasets that EDTrend which considers the possible noise or fluctuations in data, can effectively predict the continuous emergence or decayedness of patterns, and their supports in time-stamped datasets.
- M. W. K. Cheng, B. K. K. Choi, and W. K. W. Cheung. Hiding emerging patterns with local recoding generalization. In M. J. Zaki, J. X. Yu, B. Ravindran, and V. Pudi, editors, Advances in Knowledge Discovery and Data Mining, volume 6118 of LNCS, pages 158--170. Springer Berlin Heidelberg, 2010. Google ScholarDigital Library
- G. Dong and J. Li. Efficient mining of emerging patterns: Discovering trends and differences. In Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '99, pages 43--52, New York, NY, USA, 1999. ACM. Google ScholarDigital Library
- G. Dong and J. Li. Mining border descriptions of emerging patterns from dataset pairs. Knowledge and Information Systems, 8(2):178--202, 2005. Google ScholarDigital Library
- G. Dong, X. Zhang, L. Wong, and J. Li. CAEP: Classification by aggregating emerging patterns. In S. Arikawa and K. Furukawa, editors, Discovery Science, volume 1721 of LNCS, pages 30--42. Springer Berlin Heidelberg, 1999. Google ScholarDigital Library
- H. Fan and K. Ramamohanarao. An efficient single-scan algorithm for mining essential jumping emerging patterns for classification. In M. S. Chen, P. S. Yu, and B. Liu, editors, Advances in Knowledge Discovery and Data Mining, volume 2336 of LNCS, pages 456--462. Springer Berlin Heidelberg, 2002. Google ScholarDigital Library
- H. Fan and K. Ramamohanarao. Efficiently mining interesting emerging patterns. In G. Dong, C. Tang, and W. Wang, editors, Advances in Web-Age Information Management, volume 2762 of LNCS, pages 189--201. Springer Berlin Heidelberg, 2003.Google Scholar
- H. Fan and K. Ramamohanarao. Fast discovery and the generalization of strong jumping emerging patterns for building compact and accurate classifiers. IEEE Transactions on Knowledge and Data Engineering, 18(6):721--737, 2006. Google ScholarDigital Library
- R. E. Kalman. A new approach to linear filtering and prediction problems. Journal of Basic Engineering, 82 (1):35--45, 1960.Google ScholarCross Ref
- J. Li, G. Dong, and K. Ramamohanarao. Making use of the most expressive jumping emerging patterns for classification. Knowledge and Information Systems, 3(2):1--29, May 2001. Google ScholarDigital Library
- J. Li, G. Dong, K. Ramamohanarao, and L. Wong. Deeps: A new instance-based lazy discovery and classification system. Machine Learning, 54(2):99--124, 2004. Google ScholarDigital Library
- J. Li, H. Liu, J. R. Downing, A. E. J. Yeoh, and L. Wong. Simple rules underlying gene expression profiles of more than six subtypes of acute lymphoblastic leukemia (all) patients. Bioinformatics, 19(1):71--78, 2003.Google ScholarCross Ref
- J. Li and L. Wong. Emerging patterns and gene expression data. Genome Informatics, 12:3--13, 2001.Google Scholar
- J. Li, K. Ramamohanarao, and G. Dong. Combining the strength of pattern frequency and distance for classification. In D. Cheung, G. J. Williams, and Q. Li, editors, Advances in Knowledge Discovery and Data Mining, volume 2035 of LNCS, pages 455--466. Springer Berlin Heidelberg, 2001. Google ScholarDigital Library
- V. M. Nofong. Mining Productive Emerging Patterns and Their Application in Trend Prediction. In M. Z. Islam, L. Chen, K. L. Ong, Y. Zhao, R. Nayak and P. Kennedy, Thirteenth Australasian Data Mining Conference, 2015.Google Scholar
- V. M. Nofong, J. Liu, and J. Li. A study on the applications of emerging sequential patterns. In H. Wang and M. A. Sharaf, editors, Databases Theory and Applications, volume 8506 of LNCS, pages 62--73. Springer International Publishing, 2014.Google Scholar
- V. M. Nofong, J. Liu, and J. Li. Efficient Mining of Non-derivable Emerging Patterns. In M. A. Sharaf, M. A. Cheema and Jianzhong Q. editors, Databases Theory and Applications, volume 9093 of LNCS, pages 244--256. Springer International Publishing, 2015.Google Scholar
- N. Pasquier, Y. Bastide, R. Taouil, and L. Lakhal. Discovering frequent closed itemsets for association rules. In Proceedings of the 7th International Conference on Database Theory, pages 398--416. Springer-Verlag, 1999. Google ScholarDigital Library
- G. Poezevara, B. Cuissart, and B. Crèmilleux. Extracting and summarizing the frequent emerging graph patterns from a dataset of graphs. Journal of Intelligent Information Systems, 37(3):333--353, 2011. Google ScholarDigital Library
- S. Sarkka, A. Vehtari, and J. Lampinen. Time series prediction by kalman smoother with cross-validated noise density. In Proceedings of the IEEE International Joint Conference on Neural Networks, volume 2, pages 1615--1619, 2004.Google Scholar
- S. Sarkka, A. Vehtari, and J. Lampinen. Cats benchmark time series prediction by kalman smoother with cross-validated noise density. Neurocomputing, 70 (13--15):2331--2341, 2007. Google ScholarDigital Library
- H. S. Song, S. H. Kim, and J. K Kim. Mining the change of customer behavior in an internet shopping mall. Expert Systems with Applications, 21(3):157--168, 2001.Google ScholarCross Ref
- A. Soulet, B. Crèmilleux, and F. Rioult. Condensed representation of emerging patterns. In H. Dai, R. Srikant, and C. Zhang, editors, Advances in Knowledge Discovery and Data Mining, volume 3056 of LNCS, pages 127--132. Springer Berlin Heidelberg, 2004.Google Scholar
- P. Terlecki and K. Walczak. Jumping emerging patterns with negation in transaction databases - classification and discovery. Information Sciences, 177(24):5675--5690, 2007. Google ScholarDigital Library
- C. Tsai and Y. C. Shieh. A change detection method for sequential patterns. Decision Support Systems, 46(2):501--511, 2009. Google ScholarDigital Library
- G. Welch and G. Bishop. An introduction to the kalman filter. Technical Report, UNC-CH, pages TR 95--041, 2003. Google ScholarDigital Library
- J. Whittaker, S. Garside, and K. Lindveld. Tracking and predicting a network traffic process. International Journal of Forecasting, 13(1):51--61, 1997.Google ScholarCross Ref
- Y. Xie, Y. Zhang, and Z. Ye. Short-term traffic volume forecasting using kalman filter with discrete wavelet decomposition. Computer-Aided Civil and Infrastructure Engineering, 22(5):326--334, 2007.Google ScholarCross Ref
Index Terms
- EDTrend: a methodology for trend prediction with emerging and decaying patterns
Recommendations
Fast Discovery and the Generalization of Strong Jumping Emerging Patterns for Building Compact and Accurate Classifiers
Classification of large data sets is an important data mining problem that has wide applications. Jumping Emerging Patterns (JEPs) are those itemsets whose supports increase abruptly from zero in one data set to nonzero in another data set. In this ...
Observation of sales trends by mining emerging patterns in dynamic markets
One of the regular tasks for managers is to formulate various marketing strategies to confront competitive and changing market environments. Therefore, identifying product sales trends in order to evolve a better strategy is an important issue. It is ...
A Bayesian approach to use emerging patterns for classification
ADC '03: Proceedings of the 14th Australasian database conference - Volume 17Emerging Patterns (EPs) are itemsets (characteristics) whose supports change significantly from one data class to another. This work proposes a novel approach to use EPs as a basic means for classification. It is called Bayesian Classification based on ...
Comments