Gabrys, B., 2001. Data Editing for Neuro-Fuzzy Classifiers. In: Fourth International ICSC Symposium: Proceedings of the SOCO/ISFI’2001 Conference, June 26 - 29, 2001, Paisley, Scotland, p. 77.
Full text available as:
|PDF - Accepted Version|
In this paper we investigate the potential benefits and limitations of various data editing procedures when constructing neuro-fuzzy classifiers based on hyperbox fuzzy sets. There are two major aspects of data editing which we are attempting to exploit: a) removal of outliers and noisy data; and b) reduction of training data size. We show that successful training data editing can result in constructing simpler classifiers (i.e. a classifier with a smaller number and larger hyperboxes) with better generalisation performance. However we also indicate the potential dangers of overediting which can lead to dropping the whole regions of a class and constructing too simple classifiers not able to capture the class boundaries with high enough accuracy. A more flexible approach than the existing data editing techniques based on estimating probabilities used to decide whether a point should be removed from the training set has been proposed. An analysis and graphical interpretations are given for the synthetic, non-trivial, 2-dimensional classification problems.
|Item Type:||Conference or Workshop Item (Paper)|
|Subjects:||Generalities > Computer Science and Informatics > Artificial Intelligence|
Generalities > Computer Science and Informatics
|Group:||Faculty of Science and Technology|
|Deposited By:||INVALID USER|
|Deposited On:||19 Dec 2008 20:01|
|Last Modified:||10 Sep 2014 15:43|
Downloads per month over past year
|Repository Staff Only -|