The purpose of this paper is to present methods, dedicated for extract useful patterns from the eddy current measurement data. The paper presents a methodology of knowledge extraction, an association rule learning algorithm and the methods used to improve quality of the data collected by electromagnetic systems. Presented solutions were implemented in the new eddy current system used to evaluation of steel bars in reinforced concrete structures.


data analysis; association rules; eddy currents; reinforcement concrete structures

Agrawal R, Srikant R.: Fast algorithms for mining association rules in large databases, Proceedings of the 20th International Conference on Very Large Data Bases, 1994, s. 487-499.

Chady T., Frankowski P.: Electromagnetic Evaluation of Reinforced Concrete Structure, Review Of Progress in Quantitative Nondestructive Evaluation, Vol.32, 2013, s. 1355-1362.

Drobiec Ł., Jasiński R., Piekarczyk A.: Diagnostyka konstrukcji żelbetowych – Tom 1. PWN, Warszawa 2010.

Shearer C.: The CRISP-DM model: the new blueprint for data mining, J Data Warehousing, nr 5/2000, s. 13-22.

Zybura A., Jaśniok A., Jaśniok T.: Diagnostyka konstrukcji żelbetowych, Tom 2. PWN, Warszawa 2011.

Published : 2014-06-18

Frankowski, P. (2014). KNOWLEDGE EXTRACTION FROM THE EDDY CURRENT MEASUREMENT DATA. Informatyka, Automatyka, Pomiary W Gospodarce I Ochronie Środowiska, 4(2), 45-48.

Pawel Frankowski
West Pomeranian University of Technology in Szczecin  Poland