|2/2008 - 12|
Training Neural Networks Using Input Data CharacteristicsCERNAZANU, C.
|Click to see author's profile on SCOPUS, IEEE Xplore, Web of Science|
|Download PDF (710 KB) | Citation | Downloads: 742 | Views: 3,225|
neural networks, data mining, correlation-based feature subset selection method, data features extraction, training algorithm
neural(8), networks(7), data(7), selection(6), learning(6), mining(5), machine(5), ijcnn(4), feature(4)
Blue keywords are present in both the references section and the paper title.
About this article
Date of Publication: 2008-06-02
Volume 8, Issue 2, Year 2008, On page(s): 65 - 70
ISSN: 1582-7445, e-ISSN: 1844-7600
Digital Object Identifier: 10.4316/AECE.2008.02012
Web of Science Accession Number: 000264815000012
SCOPUS ID: 77955635511
Feature selection is often an essential data processing step prior to applying a learning algorithm. The aim of this paper consists in trying to discover whether removal of irrelevant and redundant information improves the performance of neural network training results. The present study will describe a new method of training the neural networks, namely, training neural networks using input data features. For selecting the features, we used a filtering technique (borrowed from data mining) which consists in selecting the best features from a training set. The technique is made up of two components: a feature evaluation technique and a search algorithm for selecting the best features. When applied as a data preprocessing step for one common neural network training algorithms, the best data results obtained from this network are favorably comparable to a classical neural network training algorithms. Nevertheless, the first one requires less computation.
|References|||||Cited By «-- Click to see who has cited this paper|
| Negnevitsky, M., "Artificial Intelligence: A Guide to Intelligent Systems", (2nd Edition), Addison Wesley, England,2005.
 Luger G., "Artificial Intelligence :Structures and Strategies for Complex Problem Solving", (Fifth Edition) Addison Wesley, 2005.
 Stergiou, C., Siganos, D., "Neural networks", [Online] Available: Temporary on-line reference link removed - see the PDF document, 1996
 Babii, S., Cretu, V., Petriu, E. M., "Performance Evaluation of Two Distributed BackPropagation Implementations", Neural Networks 2007, IJCNN 2007, pp. 1578-1583
 Zhongwen, L., Hongzhi, L., Xincai, W., "Artificial neural network computation on graphic process unit", Neural Networks, 2005, IJCNN 2005, pp. 622-626.
 Siddique, M. N. H., Tokhi, M.O., "Training neural networks: backpropagation vs. genetic algorithms", Neural Networks, 2001, IJCNN, 2001, pp. 2673-2678
 Nguyen, D., Widrow, B., "Improving the learning speed of 2-layer neural networks by choosing initial values of adaptive weights", Neural Networks 1990, IJCNN, 1990, pp. 21-26, Volume. 3
 Gorea, D., "Dynamically Integrating Knowledge in Applications. An Online Scoring Engine Architecture", Advances in Electrical and Computer Engineering, Suceava, Romania, Volume 8, 2008, pp.44-49
[CrossRef] [Full Text] [SCOPUS Times Cited 2]
 Langley, P., "Selection of relevant features in machine learning", Proceedings of the AAAI Fall Symposium on Relevance, AAAI Press, 1994
 Jain, A., Zongker, D., "Feature selection: evaluation, application and small sample performance, Pattern Analysis and Machine Learning Intelligence, IEEE Transactions on, Volume 19, 1997, pp. 153-158
 Pudil, P., Novovicova, J., Kittler, J., "Floating search methods in feature selection, Pattern Recognition Letters, Volume 15, November 1994, pp. 1119-1125.
[CrossRef] [Web of Science Times Cited 1408] [SCOPUS Times Cited 1681]
 Kim, Y., Street, W.N., Menczer, F. Roussell, G.J., "Feature selection in data mining", J. Wang Editor, Data Mining: Opportunities and Challenges, Idea Group Publishing, 2003, pages 80-105.
 Gigli, G., Bosse, I., Lampropoulos, G.A., "A optimized architecture for classification combining data fusion and data mining", Information Fusion, Volume 8, 2007, pp. 366-378
[CrossRef] [Web of Science Times Cited 5] [SCOPUS Times Cited 7]
 Hall, M. "Correlation-based Feature Selection for Machine Learning", Ph. D. diss. Hamilton, NZ: Waikato Uiversity, Department of Computer Science, 1998
 Boyan, J., Moore, A., "Learning evaluation functions to improve optimization by local search", Journal of Machine Learning Research, Volume 1, pp. 77-112, 2000
 Weka3, "Data mining Software in Java", The University of Waikato, [Online] Available: Temporary on-line reference link removed - see the PDF document, 2008
 Witten, I. H., Frank, E., "Data mining: Practical Machine Tools and Techniques", (Second Edition), Morgan Kaufmann, 2005.
 NIST Handprinted Forms and Characters Database, [Online] Available: Temporary on-line reference link removed - see the PDF document
 http://weka.sourceforge.net/wiki/index.php, Performing attribute selection, 2008
 Image Segmentation Data, Vision Group, University of Massachusetts, November, 1990.
Web of Science® Citations for all references: 1,413 TCR
SCOPUS® Citations for all references: 1,690 TCR
Web of Science® Average Citations per reference: 71 ACR
SCOPUS® Average Citations per reference: 85 ACR
TCR = Total Citations for References / ACR = Average Citations per Reference
We introduced in 2010 - for the first time in scientific publishing, the term "References Weight", as a quantitative indication of the quality ... Read more
Citations for references updated on 2017-12-13 18:35 in 26 seconds.
Note1: Web of Science® is a registered trademark of Thomson Reuters.
Note2: SCOPUS® is a registered trademark of Elsevier B.V.
Disclaimer: All queries to the respective databases were made by using the DOI record of every reference (where available). Due to technical problems beyond our control, the information is not always accurate. Please use the CrossRef link to visit the respective publisher site.
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania
All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.
Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.
Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.