Determining the significance of input parameters using sensitivity analysis
Accompanying the application of rule extraction algorithms to real-world problems is the crucial difficulty to compile a representative data set. Domain experts often find it difficult to identify all input parameters that have an influence on the outcome of the problem. In this paper we discuss the problem of identifying relevant input parameters from a set of potential input parameters. We show that sensitivity analysis applied to a trained feedforward neural network is an efficient tool for the identification of input parameters that have a significant influence on any one of the possible outcomes. We compare the results of a neural network sensitivity analysis tool with the results obtained from a machine learning algorithm, and discuss the benefits of sensitivity analysis to a neural network rule extraction algorithm.
Unable to display preview. Download preview PDF.
- [Craven 1993]MW Craven and JW Shavlik, Learning Symbolic Rules using Artificial Neural Networks, Proceedings of the Tenth International Conference on Machine Learning, 1993, pp 73–80.Google Scholar
- [Mangasarian 1990]OL Mangasarian and WH Wolberg, Cancer Diagnosis via Linear Programming, SIAM News, 23(5), September 1990, pp 1 & 18.Google Scholar
- [Prechelt 1994]L. Prechelt, PROBEN1 — A Set of Neural Network Benchmark Problems and Benchmarking Rules, Technical Report 21/94, Fakultät für Informatik, Universität Karlsruhe, Germany, September 1994.Google Scholar
- [Quinlan 1993]JR Quinlan, C4.5: Programs for Machine Learning, Morgan Kaufmann Publishers, San Mateo, California, 1993.Google Scholar
- [Theron 1994]H Theron, Specialization by Exclusion: An Approach to Concept Learning, PhD dissertation, Department of Computer Science, University of Stellenbosch, 1994.Google Scholar
- [Zurada 1993]JM Zurada, A Malinowski and I Cloete, Sensitivity Analysis for Minimization of Input Data Dimension for Feedforward Neural Network, IEEE International Symposium on Circuits and Systems, London, May 30–June 3, 1994.Google Scholar