Title :
Double mutation and correction to expand the training data space using emerging patterns
Author :
Alhammady, Hamad
Author_Institution :
Etisalat Univ. Coll., Sharjah
Abstract :
The approach of expanding the training data space has been proposed recently in the field of data mining. This approach is aimed at improving the accuracy of different classifiers. The performance of these classifiers depends on the amount of knowledge gained from the training data. The knowledge is proportional to the size of the data space. Different methods have been proposed to expand the data space (hence, the gained knowledge). In this paper, we propose a new data expansion method. We experimentally prove that our method is capable of improving the performance of a classifier more than the previous proposed methods.
Keywords :
data mining; pattern classification; classifiers; data expansion method; data mining; training data space; Data mining; Educational institutions; Genetic algorithms; Genetic mutations; Itemsets; Machine learning; Power measurement; Terminology; Training data;
Conference_Titel :
Signal Processing and Its Applications, 2007. ISSPA 2007. 9th International Symposium on
Conference_Location :
Sharjah
Print_ISBN :
978-1-4244-0778-1
Electronic_ISBN :
978-1-4244-1779-8
DOI :
10.1109/ISSPA.2007.4555445