Patterns of multirelational data transformation in data mining process

Marcin Mazurek


Multirelational data mining requires complex preprocessing of data. Identification of transformation patterns and implementation of reusable components lead to more robust data-mining flow construction process. In this paper concept of implementation of selected transformation patterns is presented. Rapid Miner envi-ronment is used to build transformations, which can be later used in predicting cus-tomer behavior.


multirelational data mining; time series; propositionalization

Full Text:

PDF (Polski)


Last M., Klein Y., Kandel A., Abraham K.: Knowledge Discovery in Time Series Databases. IEEE Transactions on Systems. Man and Cybernetics, Vol. 31, 2001, p. 160-169.

Dzeroski S.: Multirelational data mining. An introduction. ACM SIGKDD Explorations Newsletter, Vol. 5, 2003, p. 1-16.

Dzeroski S., Lavrac N.: Relational Data Mining. Springer, Berlin 2001.

Knobbe A.: Multi-Relational Data Mining. IOS Press, Amsterdam 2006.

Knobbe A., De Haas M., Siebes A.: Propositionalization and Aggregates. Proceeding of the 5th PKDD, 2001, p. 277-288.

Krogel M. A., Wrobel S.: Transformation-Based Learning Using Multirelational Aggregation. LNAI, 2001, p. 142-155.

Liao T. W.: Clustering of time series data - a survey. Pattern Recognition, Vol. 38, 2005, p. 1857-1874.

Minaei-Bidgoli B., Lajevardi S. B.: Correlation Mining between Time Series Stream and Event Stream. Networked Computing and Advanced Information Management, 2008, p. 333-338.

Morchen F.: Unsupervised pattern mining from symbolic temporal data. SIGKDD, 2006.

Rayner A.: Discovering Knowledge from Multi-relational Data Based on Information Retrieval Theory. LNAI, 2009, p. 409-416.

Sayal M.: Detecting Time Correlations In Time Series Data Streams. HP Laboratories, 2004.

Tak-Chung F.: A review on time series data mining. Engineering Application of Artificial Intelligence,2011, p. 164-181.