Comparison of recent methods for inference of variable influence in neural networks

Neural Networks

Volumn 19, Issue 4, 2006, Pages 500-513

  Papadokonstantakis, Stavros ^a   Lygeros, Argyrios ^a   Jacobsson, Sven P ^b  

^a NATIONAL TECHNICAL UNIVERSITY OF ATHENS   (Greece)

^b ASTRAZENECA R AND D   (Sweden)

Author keywords

Automatic relevance determination; General influence measure; Information theoretic approach; Sensitivity analysis; Sequential zeroing of weights; Variable influence in neural networks

Indexed keywords

DATABASE SYSTEMS; INFORMATION THEORY; MATHEMATICAL MODELS; SENSITIVITY ANALYSIS;

AUTOMATIC RELEVANCE DETERMINATION; GENERAL INFLUENCE MEASURE; INFORMATION THEORETIC APPROACH; SEQUENTIAL ZEROING OF WEIGHTS; VARIABLE INFLUENCE IN NEURAL NETWORKS;

NEURAL NETWORKS;

ACCURACY; ARTICLE; ARTIFICIAL NEURAL NETWORK; BAYES THEOREM; INFORMATION SCIENCE; INTERMETHOD COMPARISON; MATHEMATICAL COMPUTING; MATHEMATICAL MODEL; NOISE; PRIORITY JOURNAL;

EID: 33744944043     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.neunet.2005.09.002     Document Type: Article

References (20)

1. Andersson F.O., Åberg M., and Jacobsson S.P. Algorithmic approaches for studies of variable influence, contribution and selection in neural networks. Chemometrics and Intelligent Laboratory Systems 51 (2000) 61-72
2. Bollas, G. M., Papadokonstantakis, S., Michalopoulos, J., Arampatzis, G., Lappas, A. A., Vasalos, I. A., & Lygeros, A. (2003). Using hybrid neural networks in scaling up an FCC model from a pilot plant to an industrial unit. Chemical Engineering & Processing, 42(8-9), 697-713.
3. Burden F.R., Ford M.G., Whitley D.C., and Winkler D.A. Use of automatic relevance determination in QSAR studies using Bayesian neural networks. Journal of Chemical Information and Computer Sciences 40 (2000) 1423-1430
4. Hornik K., Stichcombe M., and White H. Multilayer feedforward networks are universal approximators. Neural Networks 2 (1989) 359-366
5. Howes P., and Crook N. Using input parameter influences to support the decisions of feedforward neural networks. Neurocomputing 24 (1999) 191-206
6. Kramer M.A. Nonlinear principal component analysis using autoassociative neural networks. AIChE Journal 37 (1991) 233-243
7. Lennox B., Rutherford P., Montague G.A., and Haughin C. Case study investigating the application of neural networks for process modeling and condition monitoring. Computers and Chemical Engineering 22 11 (1998) 1573-1579
8. MacKay D.J.C. Bayesian interpolation. Neural Computation 4 (1992) 415-447
9. MacKay D.J.C. A practical Bayesian framework for backpropagation networks. Neural Computation 4 (1992) 448-472
10. MacKay D.J.C. Comparison of approximate methods for handling hyperparameters. Neural Computation 11 5 (1999) 1035-1068
11. Mendenhall W., and Sincich T. Statistics for the engineering and computer sciences (1988), Dellen Publishing Company divisions of Macmillan Inc., San Francisco, CA pp. 851-855
12. Michalopoulos J., Papadokonstantakis S., Arampatzis G., and Lygeros A. Modelling of an industrial fluid catalytic cracking unit using neural networks. Chemical Engineering Research and Design 79A (2001) 137-142
13. Nord L.I., and Jacobsson S.P. A novel method for examination of the variable contribution to computational neural network models. Chemometrics and Intelligent Laboratory Systems 44 (1998) 153-160
14. Penny W.D., and Roberts S.J. Bayesian neural networks for classification: How useful is the evidence framework?. Neural Networks 12 (1999) 877-892
15. Reed R. Pruning algorithms-a survey. IEEE Transactions on Neural Networks 4 (1993) 740-747
16. Schmitz G.P.J., Aldrich C., and Gouws F.S. An algorithm for extraction of decision rules from artificial neural networks. IEEE Transactions on Neural Networks 10 6 (1999) 1390-1401
17. Sridhar D.V., Bartlett E.B., and Seagrave R.C. Information theoretic subset selection for neural network models. Computers and Chemical Engineering 22 4 (1998) 613-626
18. Thompson M.L., and Kramer M.A. Modeling chemical processes using prior knowledge and neural networks. AIChE Journal 40 8 (1994) 1328-1340
19. Weixiang Z., Dezhao C., and Shangxu H. Potential function based neural networks and its application to the classification of complex chemical patterns. Computers and Chemistry 25 2 (1998) 385-391
20. Zamarreno J.M., and Vega P. Identification and predictive control of a melter unit used in the sugar industry. Artificial Intelligence in Engineering 11 4 (1997) 365-373