A neural network based modelling and sensitivity analysis of damage ratio coefficient

The methods of extracting the contribution of variables in artificial neural network models – Comparison of inherent instability

“…Based on the error ratio values, the percentage influence of input parameters can be determined. A similar sensitivity analysis method was employed by Pastor- Barcenasa et al (2005) in surface ozone modelling and by Hadzima-Nyarko et al (2011) in the modelling of the damage ratio coefficient.…”

“…Therefore, the application of neural network committee (NNC)-based sensitivity analysis was proposed in their report in the case of geotechnical engineering problems. Hadzima-Nyarko et al (2011) have used 15 neural networks, trained by the same combination of input parameters but each time with a different weight initialization. The average network error is taken as the network error used for sensitivity analysis in the neural model of damage ratio coefficient.…”

The identification of relationships between selected honey parameters by extracting the contribution of independent variables in a neural network model

“…The errors ratio values can be used for calculating the percentage influence of input parameters. A similar SA method has been employed by HadzimaNyarko et al [44] in the modeling of the damage ratio coefficient and by Pastor-Bárcenas et al [45] in surface ozone modeling.…”

Spectral Functions of Damage Index (DI) for Masonry Buildings with Flexible Floors

“…Using a database of DI spectral functions as well as the results and expressions obtained from experiments, a program (EDABS) that relates structural dimensions and seismic loads with the dynamic properties of structures and the DI was created and presented [28]. Graphical user interface is presented in Fig.…”