Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/26161
Title: Application of an artificial neural network based sensitivity analysis technique in concreting productivity study
Authors: Lu, M
Yeung, DS
Keywords: Artificial intelligence
Backpropagation
Construction
Neural nets
Productivity
Sensitivity analysis
Issue Date: 2003
Publisher: IEEE
Source: 2003 International Conference on Machine Learning and Cybernetics, 2-5 November 2003, v. 2, p. 1207-1212 How to cite?
Abstract: As follow up effort, the present research takes advantage of the concreting productivity database as established by Lu and Anson (2004), and in particular, applies an ANN based sensitivity analysis technique to process the productivity data for the concreting operations using the stationary pump and reveal quantitative information on the significance of independent input factors. Definition of the partial derivative based sensitivity measure for a trained ANN model is distinguished for continuous, discrete, and binary variables respectively in order to factor in the effect on the model's sensitivity of the unit of measure and the relevant range of input factors of various natures, which could be considerably different in practical engineering applications. It is found that dropping the least sensitive factor identified from NN modeling brings about slight tradeoff of the prediction accuracy, while the sensitivity of the other independent factors remains relatively stable.
URI: http://hdl.handle.net/10397/26161
ISBN: 0-7803-8131-9
DOI: 10.1109/ICMLC.2003.1259670
Appears in Collections:Conference Paper

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

41
Last Week
1
Last month
Checked on Aug 20, 2017

Google ScholarTM

Check

Altmetric



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.