ファジィニューラルネットワークによる内装タイルの官能評価モデリング

Translated title of the contribution: Modeling of Sensory Evaluation for Interior Tiles Using a Fuzzy Neural Network

花井 泰三, 安藤 一徳, 野口 英樹, 本多 裕之, 高井 智代, 古橋 武, 内川 嘉樹, 小林 猛

Research output: Contribution to journalArticle

Abstract

In order to construct a novel design method for interior tiles, the relationship between image words (sensory evaluation) and total evaluation for interior tiles was studied. At first, the model for estimation of the total evaluation from image words was constructed using a fuzzy neural network (FNN) and MRA (multi regression analysis). The FNN model could estimate the total evaluation more correctly than the MRA model. Secondly, a FNN model for the estimation of the total evaluation from analytical data for interior tiles was constructed. This model was less accurate than the FNN model from image words, but could estimate within 10% errors. Lastly, in order to improve accuracy of the estimation, a FNN model for estimation of the total evaluation from both analytical data and image words was constructed. This model showed the highest accuracy among all models. "<I>Jyouhinna-gehinna</I>", "<I>shareta-yabona</I>" among image words and "average value of green color", "arithmetical mean deviation", "peak count" among analytical data were needed to construct the model for estimation of the total evaluation.
Original languageJapanese
Pages (from-to)18-23
Number of pages6
JournalKagaku Kogaku Ronbunshu
Volume24
Issue number1
DOIs
Publication statusPublished - Jan 10 1998

Fingerprint

Interiors (building)
Fuzzy neural networks
Tile
Regression analysis
Sensory analysis
Color

Cite this

ファジィニューラルネットワークによる内装タイルの官能評価モデリング. / 花井泰三; 安藤一徳; 野口英樹; 本多裕之; 高井智代; 古橋武; 内川嘉樹; 小林猛.

In: Kagaku Kogaku Ronbunshu, Vol. 24, No. 1, 10.01.1998, p. 18-23.

Research output: Contribution to journalArticle

花井泰三, 安藤一徳, 野口英樹, 本多裕之, 高井智代, 古橋武, 内川嘉樹 & 小林猛 1998, 'ファジィニューラルネットワークによる内装タイルの官能評価モデリング', Kagaku Kogaku Ronbunshu, vol. 24, no. 1, pp. 18-23. https://doi.org/10.1252/kakoronbunshu.24.18
花井泰三 ; 安藤一徳 ; 野口英樹 ; 本多裕之 ; 高井智代 ; 古橋武 ; 内川嘉樹 ; 小林猛. / ファジィニューラルネットワークによる内装タイルの官能評価モデリング. In: Kagaku Kogaku Ronbunshu. 1998 ; Vol. 24, No. 1. pp. 18-23.
@article{d332b0626a2e43da8e65b114ea6eb8ad,
title = "ファジィニューラルネットワークによる内装タイルの官能評価モデリング",
abstract = "In order to construct a novel design method for interior tiles, the relationship between image words (sensory evaluation) and total evaluation for interior tiles was studied. At first, the model for estimation of the total evaluation from image words was constructed using a fuzzy neural network (FNN) and MRA (multi regression analysis). The FNN model could estimate the total evaluation more correctly than the MRA model. Secondly, a FNN model for the estimation of the total evaluation from analytical data for interior tiles was constructed. This model was less accurate than the FNN model from image words, but could estimate within 10{\%} errors. Lastly, in order to improve accuracy of the estimation, a FNN model for estimation of the total evaluation from both analytical data and image words was constructed. This model showed the highest accuracy among all models. {"}<I>Jyouhinna-gehinna{"}, {"}<I>shareta-yabona{"} among image words and {"}average value of green color{"}, {"}arithmetical mean deviation{"}, {"}peak count{"} among analytical data were needed to construct the model for estimation of the total evaluation.",
author = "泰三 花井 and 一徳 安藤 and 英樹 野口 and 裕之 本多 and 智代 高井 and 武 古橋 and 嘉樹 内川 and 猛 小林",
year = "1998",
month = "1",
day = "10",
doi = "10.1252/kakoronbunshu.24.18",
language = "Japanese",
volume = "24",
pages = "18--23",
journal = "Kagaku Kogaku Ronbunshu",
issn = "0386-216X",
publisher = "The Society of Chemical Engineers, Japan",
number = "1",

}

TY - JOUR

T1 - ファジィニューラルネットワークによる内装タイルの官能評価モデリング

AU - 花井, 泰三

AU - 安藤, 一徳

AU - 野口, 英樹

AU - 本多, 裕之

AU - 高井, 智代

AU - 古橋, 武

AU - 内川, 嘉樹

AU - 小林, 猛

PY - 1998/1/10

Y1 - 1998/1/10

N2 - In order to construct a novel design method for interior tiles, the relationship between image words (sensory evaluation) and total evaluation for interior tiles was studied. At first, the model for estimation of the total evaluation from image words was constructed using a fuzzy neural network (FNN) and MRA (multi regression analysis). The FNN model could estimate the total evaluation more correctly than the MRA model. Secondly, a FNN model for the estimation of the total evaluation from analytical data for interior tiles was constructed. This model was less accurate than the FNN model from image words, but could estimate within 10% errors. Lastly, in order to improve accuracy of the estimation, a FNN model for estimation of the total evaluation from both analytical data and image words was constructed. This model showed the highest accuracy among all models. "<I>Jyouhinna-gehinna", "<I>shareta-yabona" among image words and "average value of green color", "arithmetical mean deviation", "peak count" among analytical data were needed to construct the model for estimation of the total evaluation.

AB - In order to construct a novel design method for interior tiles, the relationship between image words (sensory evaluation) and total evaluation for interior tiles was studied. At first, the model for estimation of the total evaluation from image words was constructed using a fuzzy neural network (FNN) and MRA (multi regression analysis). The FNN model could estimate the total evaluation more correctly than the MRA model. Secondly, a FNN model for the estimation of the total evaluation from analytical data for interior tiles was constructed. This model was less accurate than the FNN model from image words, but could estimate within 10% errors. Lastly, in order to improve accuracy of the estimation, a FNN model for estimation of the total evaluation from both analytical data and image words was constructed. This model showed the highest accuracy among all models. "<I>Jyouhinna-gehinna", "<I>shareta-yabona" among image words and "average value of green color", "arithmetical mean deviation", "peak count" among analytical data were needed to construct the model for estimation of the total evaluation.

U2 - 10.1252/kakoronbunshu.24.18

DO - 10.1252/kakoronbunshu.24.18

M3 - 記事

VL - 24

SP - 18

EP - 23

JO - Kagaku Kogaku Ronbunshu

JF - Kagaku Kogaku Ronbunshu

SN - 0386-216X

IS - 1

ER -