2reB-MDF02.dvi

Size: px

Start display at page:

Download "2reB-MDF02.dvi"

さやないそみ
5 years ago
Views:

1 SP2-B2015 Massive Data Flow Visualizing Collective Attention Using Association Networks Kazutoshi Sasahara Graduate School of Information Science, Nagoya University sasahara@nagoya-u.jp keywords: association networks, collective attention, skip-gram model, twitter, word2vec Summary The socialization of the Web changes the ways we behave both online and offline, leading to a novel emergent phenomenon called collective attention in which people s attention is suddenly concentrated on a particular reallife event. Visualizing collective attention is fundamental to understand human behavior in the digital age. Here we propose association networks to visualize usage-based, term-association patterns in a large dataset of tweets (short text messages) during collective attention events. First, we train the word2vec model to obtain vector representations of terms (words) based on semantic similarities, and then construct association networks: given some terms as seeds, the associated terms are linked with each other using the trained word2vec model, and considering the resulting terms as new seeds, the same procedure is repeated. Using two sets of Twitter data the 2011 Japan earthquake and the 2011 FIFA Women s World Cup we demonstrate how association networks visualize collective attention on these events. Provided the Japan earthquake dataset, the association networks that emerged from the most frequently used terms, such as earthquake and tsunami, exhibitdistinct network structure related to people s attention during the earthquake, whereas one that emerged from emotion-related terms, such as great and terrible, shows a large connected cluster of negative terms and small clusters of positive terms. Furthermore, we compare association networks in different datasets, using the same seed terms. These results indicate the proposed method to be a useful tool for visualizing the implicit nature of collective attention that is otherwise invisible. 1.,,.,,.,, Twitter Facebook,.,, (Computational Social Science) [Lazer 09, Miller 11]. Twitter, (Collective Attention).,,, Twitter [Lehmann 12, Sasahara 13].,.,,,.,,., (Statistical Semantics) [Furnas 83],.,,,.,,.

2 803 2.,. 2 1 Twitter Twitter 140 () (SNS). Twitter., Twitter,. Twitter, 3. Sakaki,., 3 96% Twitter [Sakaki 10]., Zhao NFL,, 90% [Zhao 11]., (, [Weng 11, Lampos 12, Panisson 14])., Takeichi, [Takeichi 14]. 2 2 Twitter Twitter,..Lehmann (), 4 [Lehmann 12].,,,..,., Sasahara,,,, JS (bit) Jan 2011 Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Date 1 (2011 ) [Sasahara 13].,,, ,2 AFC, 3 FIFA. FIFA 2011, [Sasahara 13],., 2. 3.,. word2vec,. 3 1 word2vec Mikolov word2vec, [Mikolov 13a, Mikolov 13b]. Skip-gram, word2vec. Skip-gram, w t (w t c ). 2. w t, w t+j,

3 SP2-B2015 Output w(t-2) w(t-1) w(t+1) w(t+2). 1 v w1 2 v w2 s. Projection Input 2 w(t) Skip-gram ([Mikolov 13b] )..,,.,. word2vec, 100, 200, 100 (1-of-K ) 200., () w 1,w 2,...,w T Skip-gram. 1 T log p(w t+j w t ) (1) T t=1 c j c,j 0 w v w, v w, W, (1) p(w t+j w t ). exp ( ) v w p(w O w I )= O v wl W (2) w=1 exp(v w v wl ) W (2), word2vec., word2vec,.,, [Mikolov 13a, Mikolov 13b]., Bag-of-Words LDA (Latent Dirichlet Allocation) [Blei 03],.,, word2vec. 3 2 word2vec.,, s =cos( v w1, v w2 )= v w 1 v w2 v w1 v w2 (3), w, w., s th,., w s th, N.,.,, s th, N.., w 1 w 2,, w 1 w 2. w 1,. Force-directed [Bastian 09]., word2vec, (Association Networks). s th =0.6, N =20 ( 5 6 )., s th =0.6 20,, Twitter., 10,.,,. Twitter REST API 1, , () (3)

4 805 3 ().,,,,,, (FIFA )., 610,1930, 7,497, MeCab 2,., Wikipedia 3, Mozc 4 [Sasahara 13] 5 (T T, ˆ ˆ;,,ˆoˆ,ˆ) NAIST Japanese Dictionary 5.,,,. 2 mecab/doc/index.html ,, word2vec.. Mikolov word2vec 6, Skip-gram. 200, c 5,. 5. FIFA 2011.,., 3.,. 6

5 SP2-B ,., , (1 ), (2 ), (4 ), (5 ), (8 ), (9 ) [Sasahara 13]., 6 3.,.,,.,., Skype,.,,.,.,. 5 2, 6,,,,,. 4,6.,,.,,.,., 4 (),,,,,...,,., ( CM 1 ), ( CM )., CM,.,, ,, [Sasahara 14]., 5 T T, ˆ ˆ;,, ˆoˆ, ˆ. 5, T T, ˆ ˆ;ˆ * *),. 1

6 807 5 (). T T,ˆ ˆ;,, ˆoˆ, ˆ.. 5, (),. (((( )))) ( ),,., ()., 5 ˆoˆ, 7.,,.,.,,. 5 6 Dimension 2 6 Dimension 1 7 ˆoˆ, ,.,,,,,,.,, T Tˆ ˆ;.,,. 5 5,. FIFA. 2011,,PK.,. 7(A).,, 7(A).,.,.,, 7(B).,, 2.,,.,,.

7 SP2-B2015 A B 7. (A),(B) s th. 1 s 20 (N =20), s th 0.5 6,681,., N = 200., s th 0.7,. s th = ,.. s th =0.6, N =20. 2, 9.,, ˆoˆ, T T ( )., 3.,., 2.,,3 4. 1, ()., 2, 6, T Tˆoˆ 2.,, ,,.,.,,. 1.

8 Top 20 Top # of links sth 8 s th () # of links # of association chains ( ).,, ( ) 2 1,. 10, 200 (, ).,.,,,.,, 10 ()....,,. 6.,,.,,.,,., Bag-of-Words.,.,., 11..

810 30 6 SP2-B2015 Relative freq. of users 10 0 10-1 10-2 10-3 10-4 10-5 11 10 0 10 1 10 2 10 3 # of tweets Japan earthquake 2011 FIFA Women's World Cup 2011 ().,,,.,,,..,,.,.,,.,.,,.,.. [Bastian 09] Bastian, M.

9 SP2-B2015 Relative freq. of users # of tweets Japan earthquake 2011 FIFA Women's World Cup 2011 ().,,,.,,,..,,.,.,,.,.,,.,.. [Bastian 09] Bastian, M., Heymann, S., and Jacomy, M.: Gephi: An Open Source Software for Exploring and Manipulating Networks, in Proceedings of the Third International Conference on Weblogs and Social Media, pp (2009) [Blei 03] Blei, D. M., Ng, A. Y., and Jordan, M. I.: Latent Dirichlet Allocation, The Journal of Machine Learning Research, Vol. 3, pp (2003) [Furnas 83] Furnas, G., Landauer, T., Gomez, L., and Dumais, S.: Statistical Semantics: Analysis of the Potential Performance of Keyword Information Systems, Bell System Technical Journal, Vol. 62, No. 6, pp (1983) [Lampos 12] Lampos, V. and Cristianini, N.: Nowcasting Events from the Social Web with Statistical Learning, Transactions on Intelligent Systems and Technology, Vol. 3, No. 4, pp. 72:1 72:22 (2012) [Lazer 09] Lazer, D., Pentland, A., Adamic, L., Aral, S., Barabási, A.- L., Brewer, D., Christakis, N., Contractor, N., Fowler, J., Gutmann, M., Jebara, T., King, G., Macy, M., Roy, D., and Van Alstyne, M.: Computational Social Science, Science, Vol. 323, No. 515, pp (2009) [Lehmann 12] Lehmann, J., Gonçalves, B., Ramasco, J., and Cattuto, C.: Dynamical Classes of Collective Attention in Twitter, in Proceedings of the 21st International Conference on World Wide Web, pp (2012) [Mikolov 13a] Mikolov, T., Chen, K., Corrado, G., and Dean, J.: Efficient Estimation of Word Representations in Vector Space, in Proceedings of International Conference on Learning Representations (ICLR) (2013) [Mikolov 13b] Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., and Dean, J.: Distributed Representations of Words and Phrases and their Compositionality, in Advances in Neural Information Processing Systems 26, pp (2013) [Miller 11] Miller, G.: Social Scientists Wade Into the Tweet Stream, Science, Vol. 333, No. 6051, pp (2011) [Panisson 14] Panisson, A., Gauvin, L., Quaggiotto, M., and Cattuto, C.: Mining Concurrent Topical Activity in Microblog Streams, in The 4th Workshop on Making Sense of Microposts (#Microposts2014), pp (2014) [Sakaki 10] Sakaki, T., Okazaki, M., and Matsuo, Y.: Earthquake Shakes Twitter Users: Real-Time Event Detection by Social Sensors, in Proceedings of the 19th International Conference on World Wide Web, pp (2010) [Sasahara 13] Sasahara, K., Hirata, Y., Toyoda, M., Kitsuregawa, M., and Aihara, K.: Quantifying Collective Attention from Tweet Stream, PLoS ONE, Vol. 8, No. 4: e61823 (2013) [Sasahara 14] Sasahara, K.: Quantifying Collective Mood by Emoticon Networks, in Proceedings of the 2014 ACM Web Science Conference (WebSci 14), pp , ACM (2014) [Takeichi 14] Takeichi, Y., Sasahara, K., Suzuki, R., and Arita, T.: Twitter as Social Sensor: Dynamics and Structure in Major Sporting Events, in Proceedings of the 14th International Conference on the Simulation and Synthesis of Living Systems (ALIFE14) (2014) [Weng 11] Weng, J., Yao, Y., and Leonardi, E.: Event Detection in Twitter, in Proceedings of the 5th International AAAI Conference on Weblogs and Social Media, pp (2011) [Zhao 11] Zhao, S., Zhong, L., Wickramasuriya, J., and Vasudevan, V.: Human as Real-Time Sensors of Social and Physical Events: A Case Study of Twitter and Sports Games, in Proceedings of CoRR (2011) , PD , FIRST ().

Computational Semantics 1 category specificity Warrington (1975); Warrington & Shallice (1979, 1984) 2 basic level superiority 3 super-ordinate catego

Computational Semantics 1 category specificity Warrington (1975); Warrington & Shallice (1979, 1984) 2 basic level superiority 3 super-ordinate category preservation 1 / 13 analogy by vector space Figure