PublicationsFrom: Vitaly Schetinin Date: Tue, 23 Oct 2001 13:08:01 +0200 Subject: Neural Networks Extracting Rules from EEG Data My two papers presented on NIMIA-2001 workshop are available on http://nnlab.tripod.com under Online Publications. Any feedback is welcome! Vitaly Schetinin A Neural Network Decision Tree for Learning Concepts from EEG Data Vitaly Schetinin Abstract - To learn the multi-class conceptions from the electroencephalogram (EEG) data we developed a neural network decision tree (DT), that performs the linear tests, and a new training algorithm. We found that the known methods fail inducting the classification models when the data are presented by the features some of them are irrelevant, and the classes are heavily overlapped. To train the DT, our algorithm exploits a bottom up search of the features that provide the best classification accuracy of the linear tests. We applied the developed algorithm to induce the DT from the large EEG dataset consisted of 65 patients belonging to 16 age groups. In these recordings each EEG segment was represented by 72 calculated features. The DT correctly classified 80.8% of the training and 80.1% of the testing examples. Correspondingly it correctly classified 89.2% and 87.7% of the EEG recordings. Polynomial Neural Networks Learnt to Classify EEG Signals Vitaly Schetinin Abstract - A neural network based technique is presented, which is able to successfully extract polynomial classification rules from labeled electroencephalogram (EEG) signals. To represent the classification rules in an analytical form, we use the polynomial neural networks trained by a modified Group Method of Data Handling (GMDH). The classification rules were extracted from clinical EEG data that were recorded from an Alzheimer patient and the sudden death risk patients. The third data is EEG recordings that include the normal and artifact segments. These EEG data were visually identified by medical experts. The extracted polynomial rules verified on the testing EEG data allow to correctly classify 72% of the risk group patients and 96.5% of the segments. These rules performs slightly better than standard feed-forward neural networks. ========================================== Dr. Vitaly Schetinin, Tel: x49-03641-949534 E-mail b0scvi@uni-jena.de www http://nnlab.tripod.com TheorieLabor http://www.theorielabor.de Friedrich-Schiller-Universit�t, Ernst-Abbe-Platz 4 D-07740 Jena, Germany ========================================== |
Copyright © 2001 KDnuggets. Subscribe to KDnuggets News!