Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/37514
Title: Simulation of human psychological sensory perception of clothing comfort with artifical neural network (ANN)
Authors: Wong, ASW
Li, Y 
Yeung, PKW
Lee, PWH
Keywords: Artificial intelligence
Neural networks
Sensory perception
Clothing comfort
Issue Date: 2001
Source: The 6th Asian Textile Conference : Innovation & Globalization, proceedings, Hong Kong, August 22-24, 2001 (VCD) How to cite?
Abstract: The objective of this paper is to investigate the predictability of clothing comfort from psychological sensory perceptions by using feed-forward backpropagation network in the system of Artificial Neural Network (ANN). In order to achieve the objective, a series of wear trials were conducted, in which eleven sensory perceptions (clammy, clingy, damp, sticky, heavy, prickly, scratchy, fitness, breath, thermal and comfort) were rated by twenty-two professional athletes in a controlled laboratory. They were asked to wear 4 different garments in each trial and rate the above sensory perceptions during their 90 minutes exercising period. The scores then were inputted into five different feed-forward backpropagation neural networks models, which consist of six different number of hidden and output transfer neurones. Results showed that good correlation was found between predicted and actual comfort ratings with significance at p<0.001 level with all five models, indicating the overall comfort performance was predictable by using neural networks, particularly, in the models with log sigmoid hidden neurones and pure linear output neurone. Models with a single log sigmoid hidden-layer with 15 neurones or 3 hidden layers, each with 10 log sigmoid hidden neurones are able to produce better predictions than the other models for this particular data set in the study.
URI: http://hdl.handle.net/10397/37514
ISBN: 962-367-279-7
Appears in Collections:Conference Paper

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

Page view(s)

42
Last Week
1
Last month
Citations as of Jan 8, 2018

Google ScholarTM

Check


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