近年受惠於資料探勘技術的廣泛應用，網際網路市場得以獲得新的成長動能。尤其是經由彙整社群網路使用者的情緒反應、留言關鍵字的選擇、數量、頻率與時空背景資料等，網路服務漸漸走向更為多元與客製化的型態。其中以色彩對於視覺化的應用上更是不可或缺的，目前市面上幾乎沒有任何一項產品與服務不會涉及到色彩的應用。因此色彩所帶給人情緒的影響也成為商業應用中討論的重點。

本研究使用資料探勘技術，藉由「正負情緒量詞搭配程度副詞」為關鍵字，將在Bing搜尋引擎所搜尋到的網路圖片進行色彩量化，進而討論正負面的情緒量詞「在網路圖片呈現的色彩」與「配色書的建議配色」兩者之差異。

研究設計採用各6個正負向情緒量詞，搭配9個程度副詞（含單詞），收集10本配色書於正負情緒量詞的建議配色（共216組色票），並整理以「正負情緒量詞搭配程度副詞」為關鍵字（共108項）於Bing搜尋引擎所搜尋的網路圖片結果（共2700張圖片）。

研究分析發現，由「正負情緒量詞搭配程度副詞」為關鍵字所搜尋的網路圖片經過色彩量化後，負面情緒的結果與配色書的建議配色有平均30％的一致性，正向情緒的結果與配色書的建議配色有平均20％的一致性。換言之，網路圖片搜尋結果呈現許多有別以往配色書的建議配色，例如：在網路搜尋正面情緒「快樂」的網路圖片，本應呈現配色書所建議的明亮、暖色系之色彩，但卻也會搜尋到原應屬於負面情緒的暗色、冷色系色彩等照片，而這些圖片經由色彩量化後的結果與配色書的建議配色不一致。相對地，負面情緒的不一致性則較低。

透過資料探勘技術觀察情緒色彩在網路圖片上的表現後，發現有許多網路圖片色彩量化結果與配色書建議配色不一致的地方，而這些不一致的色彩量化結果可以做為情緒色彩在產品設計使用上的新的嘗試，並可依此針對不同產業的產品或服務，提供相對應的客群偏好與反應分析以便創造產品與服務差異化。

Thanks to the recently extensive application of data mining, internet market gains a new momentum in commercial growth. In particular, by retrieving emotional reactions, selections of keywords, frequency and time-space features from users’ comments in the social media, services in the internet move to a more diversified and customized patterns. Among all of these, hues in the visualized application are indispensable. Almost any kind of product and service is relevant to it. Impacts that hues brings to human beings’ emotion, therefore, becomes a promising points in the discussion of business.

To comprehend the difference of emotionally positive and negative quantifiers between “hues of pictures shown on the internet” and “harmonious hues recommended by the color match book”, this study extracts the knowledge of data mining to quantify hues of pictures searched on Bing and set “emotionally positive and negative quantifiers along with adverbs of degree” as keywords.

There are six emotional quantifiers for the positive and the negative respectively, and each of them matches with 9 adverbs of degree. Also, 216 color samples in total are included in the study after collecting recommended harmonious hues for emotionally positive and negative quantifiers from 10 color match books. On the other hand, by applying 108 “emotionally positive and negative quantifiers along with adverbs of degree” as keywords to Bing, the study collects 2,700 pictures on the internet.

The analysis of the study shows that those quantified pictures searched by setting “emotionally positive and negative quantifiers along with adverbs of degree” as keywords on the internet reaches the consistency with an average of 30% when it comes to the negative hues recommended by the color match book. For the result of the positive hues, the rate of consistency is 20%. In other words, the quantified hues for pictures on the internet differ a lot from the recommended result in the color match book.

For instance, the hues should have been bright and warm as recommended by the color match book as a result of searching pictures with positive keyword, “Happy”, on the internet, but as a matter of fact, it leads to pictures of celebration in night club which belongs to the hues of the negative emotion, namely, darkness and cold. Obviously, after quantifying hues of pictures on the internet, the result is quite inconsistent with what is displayed from the color match book. In contrast, for pictures searched from the negative keywords, the inconsistency is relatively lower.

After applying data mining and observing patterns of emotional colors from the pictures on the internet, the result discoveries there are many inconsistencies between color match books and quantified hues of pictures on the internet. The inconsistent result creates, in fact, an opportunity to explore a new experiment where emotional hues are applied to the design of product. Moreover, according to the results of experiments and by targeting products and services in various industries, preference and reactions of customers in products and services are differentiated.

壹、中文文獻
- [韓]I.R.I色彩研究所（2010）。給設計師的專業配色圖典。北京：人民郵電出版社。
- [韓]金容淑（2011）。設計中的色彩心理學。北京：人民郵電出版社。
- ArtTone視覺研究中心（2009）。冷暖色調配色寶典。臺北市：佳魁資訊。
- ArtTone視覺研究中心（2009）。基本色配色設計典。臺北縣汐止市：博碩文化。
- CR&LF研究所（2006）。配色王：最能激發創意與觸動人心的配色技法。台北縣汐止市：博碩文化。
- Mcoo色彩研究中心（2014）。綜合技法色彩型錄。臺北市：佳魁資訊。
- 卓淑玲（2013）。台灣地區華人情緒與相關心理生理資料庫─中文情緒詞常模研究。中華心理學刊 民102, 55（4），493-523。
- 周建國 （2013）。自然色での配色手帖。臺北市：上奇資訊。
- 金日龍（2011）。色彩設計。臺北市：佳魁資訊。
- 南雲治嘉（2002）。色彩配色圖表3【範例篇】。台北市：龍溪圖書。
- 南雲治嘉（2014）。色彩配色圖表2【應用篇】。新北市永和區：龍溪圖書。
- 羅瓊鵬（2015）。漢語副詞和形容詞的程度語義研究——以“真假”組合為例。外文研究，3（3），25-31。
- 羅瓊鵬（2016）。程度、量級與形容詞“真”和“假”的語義。語言研究，36（2），94-100。
 
貳、西文文獻
- Adams, F. M. and Osgood, C. E. (1973). "A Cross-Cultural Study of the Affective Meanings of Color." Journal of Cross-Cultural Psychology. 4(2): 135-156.
- Agrawal, R., et al. (1993). Database Mining: A Performance Perspective. IEEE Transactions on Knowledge and Data Engineering, 5(6): 914-925.
- Agrawal, R., et al. (1993). Mining association rules between sets of items in large databases. ACM SIGMOD international conference on Management of data, New York, USA.
- Bloomberg, D. S. (2008). Color quantization using modified median cut. from ref: http://leptonica.net/papers/colorquant.pdf
- Cooley, R., et al. (1997). Web mining: information and pattern discovery on the World Wide Web. Proceedings Ninth IEEE International Conference on Tools with Artificial Intelligence, IEEE.
- Debevec, P. (2006). A median cut algorithm for light probe sampling. ACM SIGGRAPH 2006 Courses, Boston, MA, USA.
- Gao, X. P. and Xin, J. H. (2006). Investigation of human's emotional responses on colors. Color Research & Application, 31(5): 411-417.
- Heckbert, P. (1982). Color image quantization for frame buffer display. 9th annual conference on Computer graphics and interactive techniques, Boston, Boston, Massachusetts, USA.
- Joy, G. and Xiang, Z. (1993). Center-cut for color-image quantization." The Visual Computer, 10(1): 62-66.
- Jörgensen, C. (1998). Attributes of images in describing tasks. Information Processing and Management, 34(2-3), 161-174.
- Joshi, D., et al. (2011). Aesthetics and Emotions in Images. IEEE Signal Processing Magazine, 28(5): 94-115.
- Kim, J.-D., et al. (2003). GENIA corpus - a semantically annotated corpus for bio-textmining. Bioinformatics, 19(suppl_1): i180-i182.
- Kobayashi, S. (1981). The aim and method of the color image scale. Color Research & Application, 6(2): 93-107.
- Lee, J. and Park, E. (2011). Fuzzy Similarity-Based Emotional Classification of Color Images. IEEE Transactions on Multimedia, IEEE.
- Li, N., et al. (2015). Semi-supervised emotional classification of color images by learning from cloud. 2015 International Conference on Affective Computing and Intelligent Interaction (ACII), IEEE.
- Machajdik, J. and Hanbury, A. (2010). Affective image classification using features inspired by psychology and art theory. 18th ACM international conference on Multimedia, Firenze, Italy.
- Ou, L.-C., et al. (2004). A study of colour emotion and colour preference. Part I: Colour emotions for single colours. Color Research & Application, 29(3): 232-240.
- Ou, L.-C., et al. (2004). A study of colour emotion and colour preference. Part II: Colour emotions for two-colour combinations. Color Research & Application, 29(4): 292-298.
- Ou, L.-C., et al. (2004). A study of colour emotion and colour preference. Part III: Colour preference modeling. Color Research & Application, 29(5): 381-389.
- Oyama, T., et al. (1963). Color-affection and color-symbolism is japanese and american students. The Japanese Journal of Psychology, 34(3): 109-121.
- Phongsuphap, S. and K. Kamolrat (2015). Perceptual Colour Features for Natural Scene Image Description and Retrieval. 2015 IEEE International Conference on Systems Man and Cybernetics (SMC) Kowloon, IEEE.
- Pos, O. d. and Green-Armytage, P. (2007). Facial Expressions, Colours and Basic Emotions. Colorur: Design & Creativity, 1(1): 2, 1-20.
- Soen, T. and Shimada, T. (1987). Objective evaluation of color design. Color Research & Application, 12(4): 187-195.
- Wang, W.-N. and Yu, Y.-L. (2005). Image emotional semantic query based on color semantic description, 2005 International Conference on Machine Learning and Cybernetics(pp. 4571 - 4576). Guangzhou, China: IEEE.
- Wei-dong, C. and Wei, D. (2008). An Improved Median-Cut Algorithm of Color Image Quantization. 2008 International Conference on Computer Science and Software Engineering, IEEE.
- Zhao, S., et al. (2014). Exploring Principles-of-Art Features For Image Emotion Recognition. 22nd ACM international conference on Multimedia, Orlando, Florida,