第六章 結論與未來研究
6.3 未來研究
6.3.4 圖像中的符號意義：
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
受測者的耐性與前後一致性往往是進行問卷一大考驗，這項限制決定了問卷的結果準 確程度，未來若能採用封閉環境以變因控制之下進行
6.3.4 圖像中的符號意義
透過這次不同類別的照片群中，我們發現色彩與人物的比例分佈、亮度與彩度、身體 局部、人臉表情、或手勢是否具備符號意義，而不單單指是圖像而已，且可能具備了跨 文化與次文化的差異性。情境與符號意義背後的意涵將是未來研究可以繼續發揮的空間 這種將非文字的意義與情緒、意見、評價的關聯，可將研究議題不僅僅只限於已經存在 的文字語料中。
無論是文字、圖像、音樂、影片中的情緒偵測與內容檢索(content base IR)，仍 有許多議題值得開發研究，情緒的基本架構與分類模型將可成為相關研究的基底，透 過眾人的經驗累積，針對不同類型與時間的自然累積資料，可望建立屬於不同文化、語 言、時代下的思想與智慧，更貼近當代的思維與經驗。將一般字詞藉由情緒模型進行歸 類，也將開啟心理學、資訊科學、語言學領域新的對話空間。
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
參考文獻
[1] J. Darley, S. Glucksberg, R. Kinchla ,Psychology , Prentice Hall College Div.5th
edition ,1991.
[2] E. Fox, Emotion Science ： Cognitive and Neuroscientific Approaches to
Understanding Human Emotions. Palgrave Macmillan, Sep. 2008.
[3] R. W. Picard, E. Vyzas, and J. Healey, "Toward machine emotional intelligence ： Analysis of affective physiological state", IEEE Transactions on Pattern Analysis and
Machine Intelligence, vol. 23, no. 10, pp. 1175-1191, Oct. 2001.
[4] R. W. Picard, Affective computing. The MIT Press, 1997.
[5] P Shaver, J Schwartz, D Kirson, '' Emotion knowledge ： Further exploration of a prototype approach''. Journal of Personality and Social Psychology, Vol 52(6), 1061-1086, 1987.
[6] MJ. Power, T. Dalgleish, Cognition and emotion. From order to disorder. Hove, East Sussex, UK ： Psychology Press, 1997.
[7] C.E. Izard, The Psychology of Emotions, New York, London ： Plenum Press, 1991 [8] J.A. Russell, “A circumplex model of affect”. Journal of Personality and Social
Psychology, 39(6)：1161-1178, 1980.
[9] J. Posner, J. A. Russell, and B. S. Peterson, "The circumplex model of affect ： An integrative approach to affective neuroscience, cognitive development, and
psychopathology", Development and Psychopathology, vol. 17, no. 03, pp. 715-734,
‧
[10] R. Valitutti, "WordNet-affect：an affective extension of WordNet". Proceedings of the
4th International Conference on Language Resources and Evaluation, vol. 2004, pp.
1083-1086, 2004.
[11] B. Sigurbjörnsson and R. van Zwol, "Flickr tag recommendation based on collective knowledge". Proceeding of the 17th international conference on World Wide Web, ser.
WWW '08. New York, NY, USA
： ACM, pp. 327-336, 2008.[12] R. Jonathon Read, “Recognising affect in text using pointwise-mutual information”.
master degree thesis ,University of Sussex , 2004.
[13] R. Agrawal and R. Srikant, "Fast algorithms for mining association rules". Proc. 20th
Int. Conf. Very Large Data Bases, VLDB, pp. 487-499, 1994.
[14] M. F. Porter, "An algorithm for suffix stripping". Program, vol. 14, no. 3, pp. 130-137, 1980.
[15] D. A. Hull, "Stemming algorithms ： A case study for detailed evaluation". Journal of
the American Society for Information Science, vol. 47, no. 1, pp. 70-84, Dec. 1998.
[16] J. Xu and B. W. Croft, "Corpus-Based stemming using cooccurrence of word variants". ACM Transactions on Information Systems, vol. 16, no. 1, pp. 61-81, 1998.
[17] C. E. Osgood, "Semantic differential technique in the comparative study of cultures1,"
American Anthropologist, vol. 66, no. 3, pp. 171-200, 1964.
[18] J. Berger , Ways of Seeing. Penguin ,1972 .
[19] S. Schmidt and W. G. Stock, "Collective indexing of emotions in images. a study in emotional information retrieval", Journal of the American Society for Information
Science and Technology, vol. 60, no. 5, pp. 863-876, 2009.
[20] J. San Pedro and S. Siersdorfer, "Ranking and classifying attractiveness of photos in
‧
folksonomies”. Proceedings of the 18th international conference on World wide web,
ser. WWW '09. New York, NY, USA： ACM, pp. 771-780, 2009.
[21] J. Beaudoin, "Folksonomies： Flickr image tagging： Patterns made visible". Bul.
Am. Soc. Info. Sci. Tech., vol. 34, no. 1, pp. 26-29, 2007.
[22]鄭學侖，「以web2.0民眾分類法建置音樂推薦系統之研究」，國立政治大學資訊 管理研究所碩士論文。
[23] A.J. Gill, D.Gergle, R.M. French, and J.Oberlander, "Emotion rating from short blog texts", Proceeding of the twenty-sixth annual SIGCHI conference on Human factors in
computing systems, ser. CHI '08. New York, NY, USA
： ACM, pp. 1121-1124, 2008.[24] E. Zheleva, J. Guiver, E. M. Rodrigues, and N. M. Frayling, "Statistical models of music-listening sessions in social media", Proceedings of the 19th international
conference on World wide web, ser. WWW '10. New York, NY, USA
： ACM, pp. 1019-1028, 2010.[25] S. Siersdorfer, E. Minack, F. Deng, and J. Hare, "Analyzing and predicting sentiment of images on the social web". Proceedings of the international conference on
Multimedia, ser. MM '10. New York, NY, USA
： ACM, pp. 715-718, 2010.[26] G.Grefenstette , “Comparing the Language Used in Flickr, General Web Pages, Yahoo Images and Wikipedia”. LREC08 proceedings, 6-11, 2008.
[27] L. Kennedy, M. Naaman, S. Ahern, R. Nair, and T. Rattenbury, "How Flickr helps us make sense of the world ： context and content in community-contributed media collections". Proceedings of the 15th international conference on Multimedia, ser.
MULTIMEDIA '07. New York, NY, USA
： ACM, pp. 631-640, 2007.‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Flickr, academic article, to read". Proceedings of the seventeenth conference on
Hypertext and hypermedia, ser. HYPERTEXT '06. New York, NY, USA
： ACM, pp. 31-40, 2006.[29] G. Begelman, "Automated tag clustering ： Improving search and exploration in the tag space". Proceeding of the Collaborative Web Tagging Workshop at WWW'06, 2006.
[30] 吳筱玫、周芷伊，「Tagging的分類與知識意涵：以Flickr首頁圖片為例」，新聞學 研究，台北，2009年4月，頁265-305
[31] Mikels, J. A., Fredrickson, B. L., Larkin, G. R., Lindberg, C. M., Maglio, S. J., and Reuter-Lorenz, P. A. (2005). Emotional category data on images from the international
affective picture system. Behavior Research Methods, 37(4):626-630.
[32] Lang, P. J. (1995). The emotion probe. studies of motivation and attention. The American psychologist, 50(5):372-385.
[33] Peter Mika. (2007). “Ontologies are us: emergent semantics in folksonomy systems In
Social Networks and the Semantic Web”. International Semantic Web Conference,
International Semantic Web Conference (ISWC2005), Vol. 5, No. 1. (March 2007), pp.5-15
‧
girl 0.251149 0 0.18124 0 0.251149 0.18124 0.069909
canon 0 0 0 0.332796 0.332796 0 0.332796
light 0.150871 0.330066 0 0.339927 0.490798 0.330066 0.160732
distressing 0 0 0 0 0 0 0
water 0 0 0 2.97004 2.97004 0 2.97004
blue 0 0 0 1.69111 1.69111 0 1.69111
nikon 0 0 0 0.408218 0.408218 0 0.408218
white 0 0 0 0 0 0 0
beautiful 0.59354 0 0 1.10098 1.69452 0 1.69452
nature 0.084209 0 0 1.92086 2.00507 0 2.00507
red 0.397226 0.207607 0 0 0.397226 0.207607 0.189619
sky 0 0 0 2.16517 2.16517 0 2.16517
explore 0.475363 0 0 0.29928 0.774644 0 0.774644
clouds 0 0 0 2.41792 2.41792 0 2.41792
eyes 0 1.65201 0.651554 0 0 2.30357 -2.30357
black 0 0.847754 0.322592 0 0 1.17035 -1.17035
bw 0 0.758446 0.793352 0 0 1.5518 -1.5518
cute 0.20194 0 0.233991 1.30189 1.50383 0.233991 1.26984
color 0.721659 0 0 0.426812 1.14847 0 1.14847
reflection 0 0 0 2.92459 2.92459 0 2.92459
cat 0 0.065932 0.52268 1.67863 1.67863 0.588612 1.09002
green 0.035364 0 0 0.575884 0.611247 0 0.611247
trees 0 0 0 2.2135 2.2135 0 2.2135
abigfave 0.648464 0 0 0.828788 1.47725 0 1.47725
‧
selfportrait 0 0.978352 1.07939 0 0 2.05774 -2.05774
face 0.006291 1.35789 0.120521 0 0.006291 1.47841 -1.47211
sea 0 0 0 3.2312 3.2312 0 3.2312
sunset 0.00728 0 0 3.11273 3.12001 0 3.12001
art 0.147498 1.32063 0 0 0.147498 1.32063 -1.17313
dog 0 0 0.932046 1.16774 1.16774 0.932046 0.235694
animal 0 0.599846 0.505248 0.929607 0.929607 1.10509 -0.17549
me 0 0.462812 1.28452 0 0 1.74733 -1.74733
woman 0 0.390203 0.16872 0 0 0.558923 -0.55892
beach 0 0 0 2.79941 2.79941 0 2.79941
people 1.09578 0 0 0.257064 1.35284 0 1.35284
smile 3.10134 0 0 0 3.10134 0 3.10134
fun 3.01822 0 0 0 3.01822 0 3.01822
peace 0 0 0 3.56617 3.56617 0 3.56617
sun 0.373277 0 0 2.19861 2.57189 0 2.57189
self 0 0.628268 1.19844 0 0 1.82671 -1.82671
blackandwhite 0 0.869802 0.389064 0 0 1.25887 -1.25887
landscape 0 0 0 3.37907 3.37907 0 3.37907
hand 0.180117 0.9257 0.116764 0 0.180117 1.04246 -0.86235
bokeh 0.798438 0 0 0 0.798438 0 0.798438
boy 1.11008 0.091334 0 0 1.11008 0.091334 1.01875
aplusphoto 0.448568 0 0 1.65497 2.10354 0 2.10354
flower 0.917079 0.670774 0.035168 0 0.917079 0.705941 0.211138
photography 0.783765 0 0 0 0.783765 0 0.783765
summer 0.617856 0 0 1.53637 2.15422 0 2.15422
yellow 0.386144 0 0 0.401145 0.787289 0 0.787289
photoshop 0 1.81506 0.211603 0 0 2.02666 -2.02666
night 0 0.866641 0 0.885078 0.885078 0.866641 0.018437
anawesomeshot 0.505042 0 0 1.66738 2.17243 0 2.17243
‧
mywinners 0.279508 0 0 1.40171 1.68122 0 1.68122
man 0 0.698873 0.757838 0 0 1.45671 -1.45671
pink 0.554936 0 0 0 0.554936 0 0.554936
child 1.6668 0.082433 0.094069 0 1.6668 0.176501 1.49029 shadow 0 1.19523 0.063578 0.141318 0.141318 1.25881 -1.11749
lake 0 0 0 3.45922 3.45922 0 3.45922
baby 0.756608 0 0.096415 0.986749 1.74336 0.096415 1.64694
colour 0.707133 0 0 0.841599 1.54873 0 1.54873
old 0 1.67509 0.812537 0 0 2.48763 -2.48763
orange 0 0 0 1.16728 1.16728 0 1.16728
travel 0.431155 0 0 1.87251 2.30366 0 2.30366
street 0.115576 0.263306 0.787288 0 0.115576 1.05059 -0.93502
heart 2.24759 0 0 0 2.24759 0 2.24759
emotion 0.629736 1.74689 0.527128 0 0.629736 2.27402 -1.64428
texture 0 2.14777 0.029382 0 0 2.17715 -2.17715
365 0.83729 0.587541 0.726549 0 0.83729 1.31409 -0.4768 life 0.660911 0.262583 0 0 0.660911 0.262583 0.398328
macro 0.686639 0 0 0 0.686639 0 0.686639
morning 0 0 0.165665 2.1848 2.1848 0.165665 2.01914
hdr 0 0 0 2.29879 2.29879 0 2.29879
photo 0.46584 0.096781 0 0 0.46584 0.096781 0.369059
hair 0 0.341893 0.050096 0 0 0.391989 -0.39199
ocean 0 0 0 3.1084 3.1084 0 3.1084
50mm 0.237918 0 0.221234 0 0.237918 0.221234 0.016684
film 0 0.47848 0.265682 0 0 0.744162 -0.74416
bird 0 1.50199 0 0 0 1.50199 -1.50199
friends 1.61744 0 0 0 1.61744 0 1.61744
winter 0.095255 0 0.009521 1.26662 1.36188 0.009521 1.35235
blueribbonwinner 0.473848 0 0 1.57023 2.04407 0 2.04407
‧
rocks 0 0 0 3.05826 3.05826 0 3.05826
funny 0.32964 0.667711 0.086712 0.274825 0.604465 0.754423 -0.14996 dream 0 0.949575 0 0.278705 0.278705 0.949575 -0.67087
supershot 0.786438 0 0 1.40611 2.19255 0 2.19255
impressedbeauty 0.712833 0 0 0.249507 0.96234 0 0.96234
bed 0 0 1.51754 1.19964 1.19964 1.51754 -0.3179
vintage 0 2.44745 0 0 0 2.44745 -2.44745
platinumphoto 0.516101 0 0 1.40163 1.91773 0 1.91773
goldstaraward 0.766382 0 0 1.11268 1.87906 0 1.87906
holiday 0.964552 0 0 1.62004 2.58459 0 2.58459
joy 3.86821 0 0 0 3.86821 0 3.86821
soe 0.580324 0 0 1.70182 2.28215 0 2.28215
alone 0 0.847276 1.43217 0 0 2.27944 -2.27944
day 1.0675 0 0 0 1.0675 0 1.0675
wood 0 2.00374 0 0.348754 0.348754 2.00374 -1.65499
pet 0 0 0.872378 1.48272 1.48272 0.872378 0.610342
sweet 1.12455 0 0 0.498843 1.62339 0 1.62339
explored 0.405 0 0.314664 0 0.405 0.314664 0.090336
eos 0 0.45684 0 0 0 0.45684 -0.45684
mountain 0 0 0 2.98097 2.98097 0 2.98097
Flickrdiamond 0.892252 0 0 1.46669 2.35894 0 2.35894
cold 0 0 0.62549 0.890173 0.890173 0.62549 0.264683
expression 0.817763 1.40768 0 0 0.817763 1.40768 -0.58991
paint 0 2.02544 0 0 0 2.02544 -2.02544
window 0 0.736786 1.00645 0 0 1.74323 -1.74323
snow 0.072887 0 0 1.68548 1.75837 0 1.75837
dof 0.322671 0 0.409144 0 0.322671 0.409144 -0.08647
kid 1.47852 0 0.12359 0 1.47852 0.12359 1.35493
city 0.599637 0 0 0.26561 0.865247 0 0.865247
‧
children 2.62448 0 0 0 2.62448 0 2.62448
jump 3.48626 0 0 0 3.48626 0 3.48626
new 1.71733 0 0 0 1.71733 0 1.71733
365days 0.048275 0.817385 1.24019 0 0.048275 2.05757 -2.0093
sand 0 0 0 3.21276 3.21276 0 3.21276
boat 0 0 0 3.17965 3.17965 0 3.17965
sunrise 0 0 0 3.03761 3.03761 0 3.03761
Flickr 0.717431 0 0.056654 0.268158 0.985589 0.056654 0.928935 interestingness 1.02689 0 0.138111 0 1.02689 0.138111 0.888783
couple 2.04992 0 0 0 2.04992 0 2.04992
vacation 0 0 0 2.43654 2.43654 0 2.43654
park 0.761688 0 0 1.824 2.58569 0 2.58569
silhouette 0 0.467716 0 1.19141 1.19141 0.467716 0.723694
bravo 1.52406 0 0 0.493756 2.01782 0 2.01782
digital 0.07729 0.778027 0 0 0.07729 0.778027 -0.70074
grass 0.127707 0 0 1.05611 1.18382 0 1.18382
kitty 0 0 0.035453 1.91549 1.91549 0.035453 1.88004
waves 0.289511 0 0 2.48723 2.77674 0 2.77674
lonely 0 0 2.41855 0 0 2.41855 -2.41855
family 2.03028 0 0 0 2.03028 0 2.03028
colorful 1.65126 0 0 0.135575 1.78684 0 1.78684
puppy 0 0 1.30252 1.57259 1.57259 1.30252 0.27007
scary 0 3.84374 0 0 0 3.84374 -3.84374
kitten 0 0 0.689302 1.97131 1.97131 0.689302 1.28201 brown 0 0.632273 0.005541 0.38808 0.38808 0.637814 -0.24973
spring 0.682383 0 0 1.09464 1.77702 0 1.77702
usa 1.12101 0.176115 0 0.450219 1.57123 0.176115 1.39511
kids 2.80106 0 0 0 2.80106 0 2.80106
platinumheartaward 0.278776 0 0 1.92307 2.20184 0 2.20184
‧
kiss 2.39894 0 0 0 2.39894 0 2.39894
strobist 0.451399 1.91062 0.287416 0 0.451399 2.19804 -1.74664
island 0 0 0 3.08665 3.08665 0 3.08665
california 0.737703 0 0 0.805975 1.54368 0 1.54368
urban 0 1.02319 0.14622 0 0 1.16941 -1.16941
river 0 0 0 3.07795 3.07795 0 3.07795
square 0 0.382512 0.14622 0.307839 0.307839 0.528732 -0.22089 closeup 0.112407 0.084192 0.261881 0 0.112407 0.346073 -0.23367
uk 0 0.102114 0 1.25664 1.25664 0.102114 1.15453
look 0.282474 1.50559 0.249179 0 0.282474 1.75477 -1.4723
play 3.14914 0 0 0 3.14914 0 3.14914
young 1.22016 0.014543 0.243144 0 1.22016 0.257687 0.962471
cry 0 1.33737 2.09329 0 0 3.43066 -3.43066
2009 0.920812 0 0 0 0.920812 0 0.920812
cool 0.906077 0.073545 0 0.473788 1.37987 0.073545 1.30632 rain 0.430779 0.336579 0.903386 0 0.430779 1.23996 -0.80919 autumn 0 0 0.498559 1.68209 1.68209 0.498559 1.18353
wow 1.37952 0 0 0 1.37952 0 1.37952
fall 0.274191 0.244908 0 0.492163 0.766354 0.244908 0.521446
home 0 0.336086 0.649595 0 0 0.985681 -0.98568
yawn 0 0 1.91308 1.88058 1.88058 1.91308 -0.0325
music 1.40872 0 0 0 1.40872 0 1.40872
pretty 1.49518 0 0 0 1.49518 0 1.49518
2010 0.868078 0.302052 0 0 0.868078 0.302052 0.566026 naturesfinest 0.601436 0.234218 0 1.59667 2.19811 0.234218 1.96389
longexposure 0 0 0 2.91906 2.91906 0 2.91906
dress 0.844103 0.085436 0.228731 0 0.844103 0.314166 0.529936 purple 0.254809 0 0.40864 0.631032 0.885841 0.40864 0.477201
amazing 1.27926 0 0 0.837146 2.1164 0 2.1164
‧
house 0 1.71147 0 0.155659 0.155659 1.71147 -1.55581evening 0 0 0 2.42211 2.42211 0 2.42211
adorable 0.586625 0 0.271934 1.17804 1.76467 0.271934 1.49273
i 0.829807 0 0.7082 0 0.829807 0.7082 0.121607
rest 0 0 1.99269 1.52124 1.52124 1.99269 -0.47145
interesting 0.730076 0 0 0 0.730076 0 0.730076
nikkor 0 0 0.199145 0.591625 0.591625 0.199145 0.39248 italy 0.31729 0.513686 0.133004 0.722869 1.04016 0.64669 0.393469
pain 0 3.76269 0.917275 0 0 4.67997 -4.67997
death 0 3.03569 0.728235 0 0 3.76393 -3.76393
forest 0 0.949523 0 1.58169 1.58169 0.949523 0.632166 glass 0.129471 0 0.842362 0.116466 0.245937 0.842362 -0.59643
horror 0 3.17659 0 0 0 3.17659 -3.17659
soft 0 0 0 1.07786 1.07786 0 1.07786
still 0 0 0 3.06086 3.06086 0 3.06086
serene 0 0 0 4.28148 4.28148 0 4.28148
toy 0.364794 0.999216 0.358713 0 0.364794 1.35793 -0.99314
feet 0 0 1.41813 1.21624 1.21624 1.41813 -0.20189
time 0.494468 0 0.509944 0 0.494468 0.509944 -0.01548
wall 0 2.1688 0.451708 0 0 2.6205 -2.6205
teeth 0.21436 2.16286 0 0.017878 0.232238 2.16286 -1.93062
design 0 2.12123 0 0 0 2.12123 -2.12123
tears 0 0.703922 2.18879 0 0 2.89271 -2.89271
book 0.303229 0.864465 0 0.564418 0.867647 0.864465 0.003182 india 0 0.500278 0.091403 0.097852 0.097852 0.591681 -0.49383
colorphotoaward 0.916336 0 0 2.01045 2.92678 0 2.92678
fog 0 0.64201 0.091403 2.09785 2.09785 0.733413 1.36444
decay 0 3.12642 0.614965 0 0 3.74138 -3.74138
europe 0.577136 0 0 1.73049 2.30763 0 2.30763
‧
waiting 0.451399 0 2.09058 0 0.451399 2.09058 -1.63918male 1.15184 0.695257 0 0.338858 1.4907 0.695257 0.795441
quiet 0 0 0 3.75032 3.75032 0 3.75032
feeling 0 1.33938 0.868833 0 0 2.20821 -2.20821
nap 0 0 1.83544 1.97541 1.97541 1.83544 0.13997
england 0 0.69469 0 1.18625 1.18625 0.69469 0.49156
picture 0.937913 0 0 0 0.937913 0 0.937913
christmas 2.60674 0 0 0 2.60674 0 2.60674
candid 0 0.616687 2.04052 0 0 2.65721 -2.65721
abandoned 0 2.98701 1.09104 0 0 4.07805 -4.07805
camera 1.36 0 0.530318 0 1.36 0.530318 0.829685
canada 0.186054 0 0 1.19774 1.38379 0 1.38379
scream 0.877932 3.76369 0 0 0.877932 3.76369 -2.88576
garden 0.927497 0 0 0.77698 1.70448 0 1.70448
d80 0 0 0.039406 0.931863 0.931863 0.039406 0.892457
fly 1.91314 0.241238 0 0 1.91314 0.241238 1.6719
mad 0 3.0195 0 0 0 3.0195 -3.0195
geotagged 0.416566 0 0 1.34023 1.75679 0 1.75679
Flickrsbest 0.821771 0 0 0.954166 1.77594 0 1.77594
female 1.2994 1.28972 0.612378 0 1.2994 1.9021 -0.6027
2008 1.46538 0 0 0 1.46538 0 1.46538
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
附錄B 問卷圖片與分析
‧
sum_n 2.11 polarity 4.72 社群數值 留言數 186觀看數 17426 加入最愛 335 收藏/觀看 0.02 頁面連結 ： http：//www.Flickr.com/photos/ronaldo_f_cabuhat/4906676724/
原始tag Summer Wedding,The Edison Club,Rexford, NY,View photos from this member or from everyone,Canon EOS 5D Mark II,Canon EF 24-105mm f/4 L IS USM,bride and groom,maid of honor,flowergirl,matron of honor,bridesmaid,wedding, bride,groom,summer in new york,summer colors,Cabuhat,wedding entourage, interesting,lovers,friends,love,promise,beauty,
forever,bestfriends,depth,gowns,wedding gown,hills,best friends forever,BFF,lifetime partners,couple,husband and wife, precious moment,group picture,colors,flowers, shades,light,patio,veranda,terrace, door,windows,happy,happiness,joy,true love,soulmate,
family,husband,partner,picnik,photography,image,scene,scenery,arts,Canon 580EX II,red dress,New York,red, laughter,friendship,mansion,home,house,creative photography,
wedding photography, kiss,kissing,sweet,wedding portrait,portrait (77)
過濾後 interest,friend,love,beauti,coupl,color,flower,light,window,happi,joi,famili,photography,art,red,h ome,hous,kiss,sweet,portrait (20)
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧
what where when who
原始 35 7 1 3 4 3 0 3 5 2 6 8 77
過濾 9 0 0 2 6 0 0 0 0 1 2 0 20
這張照片為結婚的場景，標籤描述畫面中的人時地物訊息，直接指涉畫面中的時空 背景與人物關係，例如 kiss,door,windows,maid of honor,flower girl,summer wedding,couple,red dress,family,group picture等等，其次主要是事物描述，例如 sweet,portrait,color,couple
等主觀的詮釋，無法從影像直接解讀的意義，則歸類為感受,評價與聯想，例如 love,happy,soulmate,friendship,forever,promise，相機資訊,個人分類,相片集則 比較少數，如 weddingphotography,weddingportrait, canon580exii,
canoneos5dmarkii。
經過stemming過濾合併字詞後，剩下的標籤僅剩人時地物,事物描述與評價,感受，
佔原始標籤總數的27%，保留大多數的內容資訊。
‧
sum_p 9.96
sum_n 0.09
polarity 9.87 社群
頁面連結 ： http：//www.Flickr.com/photos/got2bme/3613994649/
原始tag girls,beach,jump,sand,kim,brimhall,lily,blue,action,canon,just,wanna,have,fun ,water,swimsuits,50d,summer (18)
過濾後 stem
girl,beach,jump,sand,blue,canon,fun,water,summer (9)
過濾比例 0.5
what where when who
原始 4 1 1 2 4 2 0 0 0 4 1 0 18
過濾 4 0 1 0 2 1 0 0 0 0 1 0
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
這張在海灘上跳躍的照片，可以透過原始的標籤直接映射內容，主要仍以人時地物, 事物描述為主，如女孩,海灘,跳躍動作,沙灘,與地名brimhall ,藍天等等。另外有兩個 相機資訊標籤,個人評價的部份作者將一句話「just,wanna,have,fun,water」拆成多個 tag，其中還帶有感受標籤 fun。經過stemming，地點與人名被過濾掉，留下時間與物 件的標籤，以及事物描述blue,jump，和評價標籤fun，相機品牌canon仍被保留了下來，
透過stemming與詞頻過濾可保留50%標籤。
‧
sum_p 14.09 sum_n 6.66 polarity 7.43 社群 頁面連結 ： http：//www.Flickr.com/photos/todorrovic/2287792473/
原始tag dog,puppy,red,black,look,eyes,adorable,animal,cute,lovely,doggie,pup,blackie,dogs,portrait,swe et,animals,perfect,precious,friend,expression,pet,face,fabulous,gorgeous,fantastic,blackdog,hot moment,pretty,tender,view,little,beautiful,love,play,breathtaking,beograd,belgrade,serbia,srbija, blanket,warm,heart,texture,ear,baby,macro,happyholidays,happynewyear,iloveyou,sweetheart,c anis,pointer,dogbreeds,hot,aww,new,top,nostalgia,dream,dolce,cane,preatty,singidunum,canine, nose,family,wow,intense,smile,cool,innocent,mystery,indoors,puppies (75)
過濾後 stem
dog,puppi,red,black,look,ey,ador,anim,cute,love,portrait,sweet,friend,express,pet,face,pretti,bea uti,plai,heart,textur,babi,macro,new,dream,famili,wow,smile,cool (29)
過濾比例 0.39
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧
what where when who原始 17 5 0 2 13 1 0 1 4 13 5 14 75
mystery,sweet,iloveyou, happyholidays,warm, perfect,precious, fantastic 這些主觀的形容字詞，可能是一般觀者無法直接解讀的訊息。另外透過不同的標籤 了地名訊息被拿掉之外，仍可透過stem詞幹解讀出本來tag list的部份原意，透過 stemming與詞頻過濾可保留39%標籤。
‧ Q2 負向高強度「our method」
來源字詞scary sum_n 10.37 polarity
-10.21
頁面連結 ： http：//www.Flickr.com/photos/rosie_hardy/2942438120/
原始tag fear,phobia,hand,clown,cry,point,shadow,crouch,alone,lonely,terrified,scared,scary,mind,control ,lost (16)
過濾後 stem
fear,hand,cry,shadow,alon,lone,scare,scary (8)
過濾比例 0.5 what where when who
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
原始 2 0 0 0 4 0 0 0 0 1 7 2 16
過濾 1 0 0 0 1 0 0 0 0 0 6 0 8
作者使用 phobia,fear,alone,lonely,terrified與scared描述該照片的害怕與孤 單的情境，手部剪影,畫面中人物的表情與姿勢都有標籤描述，lost,control與mind則 歸類在評價與聯想無法從畫面中解讀的延伸意義。
經過stemming過後，保留下情緒相關的形容字詞fear,cry,scary,alone,lone與 scary，描述畫面內容僅剩下hand與shadow。評價與聯想的標間則被過濾掉了。透過 stemming與詞頻過濾可保留50%標籤。
‧
sum_p 0.72 sum_n 2.61 polarity -1.9 社群 頁面連結 ： http：//www.Flickr.com/photos/noamg/175696015/
原始tag noam,galai,self,portrait,black,white,bw,scream,shout,yell,screaming,noamgalai,
יאלגםעונ,newyork,nikond70,anger,blackandwhite,,העע,,הלרו,ק,יאלג,םעונ,ףוריט,העוע,קעוצ,ה,רצ,הקעצ עגוקמ,syco,psyco,mouth,up,backlight,light,talion,disperation,desperation,hate,hurt,bad,abuse,p1f 1,aplusphoto,wow,11207,supershot,wwwnoamgalaicom,fear,yelling,shouting,afraid,phobias,phot ography,allrightsreserved,photomania,,רוצ,selfportrait,photo,picture,
יעוצקמםלצ,gritar,photograph,noamg,םוליצ,הנומת,grito,ةخرص,gritando,
خارص,cridar,schrei,screamography,libertad,manifestació,thestolenscream,cultural icon (75)
過濾後 stem
self,portrait,black,white,bw,scream,angri,light,wow,fear,photography,pictur (12)
過濾比例 0.16
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧
what where when who原始 36 1 0 6 4 1 1 0 10 3 8 5 75
過濾 2 0 0 1 6 0 0 0 1 0 2 0 12
由於有部分阿拉伯文與西班牙語標籤：אםע םעונ(Noam Galai攝影者名),cridar(吶 喊),schrei(吶喊),خارص(尖叫), עעושמ(瘋狂), סעכ(憤怒),ןבם רוחש(黑白), םעונ(愉 悅),ףוריט(瘋狂),סססס (憤怒), קעוצ(喊), החרצ(驚叫), הקעצ(哭), יעוצקמםםצ(專業 攝影), םוםיצ הנומת(照片),ةخرص(哭),خارص,(尖叫),grito(西班牙語的哭泣),schrei/, 尖叫(德文)。另外 bad,talion,abuse一字帶有評價意謂,aplusphoto則為攝影社群間的 標記,作者用了至少三種語言描述該張張照片的情緒資訊，但非英語的部分都會被本實 驗stemming階段過濾，因此stemming後僅保留angri與fear兩個感受字詞，黑白與人像 與尖叫的內容，也都能夠保留下來，雖然過濾比例只有原本字詞的16%，但不置於偏離 原意太多。
‧
sum_n 4.61 polarity -4.38 社群數值 留言數 275頁面連結 ： http：//www.Flickr.com/photos/philkneen69/4558355565/
原始tag rubber band face,canon eos 5d mk2,24105l,canon,aperture 3,silver efex pro,nik,portrait,face,pain,eyes,teeth,nose,iso400,marginsquarecom (16)
過濾後 stem
canon,portrait,face,pain,ey,teeth (6)
過濾比例 0.4 what where when who
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
原始 7 0 0 1 0 4 1 0 2 0 1 0 16
過濾 4 0 0 0 0 1 0 0 0 0 1 0 6
這張照片的標籤資訊缺乏評價與聯想標籤，也沒有針對細節特徵進行描述，作者僅 針對內容,相機資訊,社群與感受進行標籤，經過stemming過濾後，仍可保留影像內容 與少數的感受與相機資訊，由於作者下標籤使用了大多數人常用的字詞，因此在 stemming階段留下了40%的標籤，所以綜合指標分數打的折扣比較少，應該是這張照片 脫穎而出的關鍵。另外社群資訊中的留言數,觀看次數與加入最愛的次數是讓這張照片 被浮現的另一個因素。從影像內容看來，的確是比較不尋常的臉部特寫。
‧
sum_n 3.63 polarity -2.23 社群數值 留言數 200頁面連結 ： http：//www.Flickr.com/photos/aihibed/1410398234/
原始tag child,sad,cry,clown,aihibed,photography,lookagain (7)
過濾後 stem
child,sad,cry,photography (4)
過濾比例 0.57 what where when who
[註]留言、觀看、加入最愛與收藏觀看比，皆大於抽樣平均
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
原始 3 0 0 1 0 0 0 0 1 0 2 0 7
過濾 2 0 0 0 0 0 0 0 0 0 2 0 4
這張低頭悲傷的白臉小丑，作者僅下了七個標籤，其中有兩個描述感受情緒，一個 人分類，其餘皆為內容描述，經過詞幹過濾之後，仍能留下超過50%的標籤，讓其綜合 指標打折扣的比例較少，人物臉部與表情的特寫較能直接連結的情緒表達。
‧
sum_n 4.08 polarity -0.31 社群數值 留言數 505頁面連結 ： http：//www.Flickr.com/photos/statelibraryofnsw/2959326615/
原始tag husky,huskie,puppy,mask,dog,snow,antarctica,blizzard,pup,ice,bw,grey,eyes,cute,frankhurley,hu rley,animal,sled dog,welpe,huskies,pups,adorable,lonely,black,white,
photography,picture,furry,state library of news outh wales,canine,mammal, 1911-1914,magical beauty,bonito,sweetie (35)
過濾後 stem
puppi,dog,snow,bw,ey,cute,anim,ador,lone,black,white,photography,pictur (13)
過濾比例 0.36
1 2 3 4 5 6 7 8 9
人時地物 事物 相機 社群 相片 個人 評價 感受 聯想 總數
人時地物 事物 相機 社群 相片 個人 評價 感受 聯想 總數