A Comparison of Two Contributive Analysis Methods Applied to an ANN Modeling Facial Attractiveness

Artificial neural networks (ANNs) are powerful predictors. ANNs, however, essentially function like 'black boxes' because they lack explanatory power regarding input contribution to the model. Various contributive analysis algorithms (CAAs) have been developed to apply to ANNs to illuminate the influences and interactions between the inputs and thus, to enhance understanding of the modeled function. In this study two CAAs were applied to an ANN modeling facial attractiveness. Conflicting results from these CAAs imply that more research is needed in the area of contributive analysis and that researchers should be cautious when selecting a CAA method

[1]  Uygar Özesmi,et al.  An artificial neural network approach to spatial habitat modelling with interspecific interaction , 1999 .

[2]  Derek Partridge,et al.  Efficient and Effective Feature Selection in the Presence of Feature Interaction and Noise , 2001, ICAPR.

[3]  Martin F. Lambert,et al.  A Statistical Input Pruning Method for Artificial Neural Networks Used in Environmental Modelling , 2004 .

[4]  W. Wang,et al.  A Comparative Study of Feature-Salience Ranking Techniques , 2001, Neural Computation.

[5]  Karen Joy,et al.  An Artificial Neural Network Model of Facial Attractiveness , 2006 .

[6]  Julian D. Olden,et al.  Illuminating the “black box”: a randomization approach for understanding variable contributions in artificial neural networks , 2002 .

[7]  Fay Sudweeks,et al.  Typicality in computer mediated discussions-an analysis with neural networks , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[8]  Michele Scardi,et al.  Developing an empirical model of phytoplankton primary production: a neural network case study , 1999 .

[9]  G. David Garson,et al.  Interpreting neural-network connection weights , 1991 .

[10]  Walter Dosch,et al.  Proceedings of the Second international conference on Software Engineering Research, Management and Applications , 2004 .

[11]  M. Gevrey,et al.  Review and comparison of methods to study the contribution of variables in artificial neural network models , 2003 .

[12]  I. Dimopoulos,et al.  Application of neural networks to modelling nonlinear relationships in ecology , 1996 .

[13]  Audrey Chen,et al.  Brain asymmetry and facial attractiveness: Facial beauty is not simply in the eye of the beholder , 1997, Neuropsychologia.

[14]  Richard Coyne,et al.  Computer Assistance in Designing from Precedent , 1992 .

[15]  Russell G. Death,et al.  An accurate comparison of methods for quantifying variable importance in artificial neural networks using simulated data , 2004 .

[16]  D. Symons.  Beauty is in the adaptations of the beholder: The evolutionary psychology of human female sexual attractiveness. , 1995 .