| Abstract |
| This study examined the emotional structure of identically toned two-color combinations in naturally dyed silk fabrics and developed an artificial intelligence-based prediction model. A total of 100 two-color combinations were produced using various natural dyes under different dyeing conditions, and their color emotions were evaluated through expert-based subjective assessments. Basic statistical analyses were conducted to examine the effects of identical-tone types, chromatic versus achromatic tones, physical color characteristics, and quantitative color-combination variables affected factors related to color emotions. Four color emotional factors―Pleasant, Classical, Soft, and Modern―were identified; these factors varied by tone types and chromaticity and were correlated with objective parameters including color variables of single colors and two-color combinations respectively. Random Forest-based prediction model were established for each color emotional factor, with higher performance observed for “Pleasant” and “Soft.” Moreover, feature importance based on SHapley Additive exPlanations (SHAP) values and surrogate linear models to identify affective color variables for emotion factor. Significant correlations between observed and predicted values indicate that the proposed artificial intelligence approach of using Random Forest models can be effectively applied to predicting color emotions in naturally dyed textiles. |
|
|
| Key Words |
| Artificial Intelligence Model, Machine Learning, Color Emotion, Two-Color Combination, Natural Dyeing, 인공지능모델, 기계학습, 색채감성, 2-배색, 천연염색 |
|
|
 |
|
