Protest event analysis is an important method for the study of collective action and social movements and typically draws on traditional media reports as the data source. We introduce collective action from social media (CASM)—a system that uses convolutional neural networks on image data and recurrent neural networks with long short-term memory on text data in a two-stage classifier to identify social media posts about offline collective action. We implement CASM on Chinese social media data and identify more than 100,000 collective action events from 2010 to 2017 (CASM-China). We evaluate the performance of CASM through cross-validation, out-of-sample validation, and comparisons with other protest data sets. We assess the effect of online censorship and find it does not substantially limit our identification of events. Compared to other protest data sets, CASM-China identifies relatively more rural, land-related protests and relatively few collective action events related to ethnic and religious conflict.
AbadiMartínAgarwalAshishBarhamPaulBrevdoEugeneChenZhifengCitroCraigCorradoGreg S., et al. 2015. “TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems.” Software available from tensorflow.org.
2.
AdamsNicholas. 2014. “Researchers to Crowds to Algorithms: Building Large, Complex, and Transparent Databases from Text in the Age of Data Science.” SSRN Scholarly Paper ID 2459325, Social Science Research Network, Rochester, NY.
3.
AdaySeanFarrellHenryLynchMarcSidesJohnFreelonDeen. 2012. “New Media and Conflict after the Arab Spring.” Washington: United States Institute of Peace80:1–24.
4.
AlmeidaPaul D.2003. “Opportunity Organizations and Threat-Induced Contention: Protest Waves in Authoritarian Settings.” American Journal of Sociology109:345–400.
5.
AlmeidaPaul D.LichbachMark. 2003. “To the Internet, from the Internet: Comparative Media Coverage of Transnational Protests.” Mobilization: An International Quarterly8:249–72.
6.
AzarEdward E.CohenStanley H.JukamThomas O.McCormickJames M.1972. “The Problem of Source Coverage in the Use of International Events Data.” International Studies Quarterly16:373–88.
7.
BahdanauDzmitryChoKyunghyunBengioYoshua. 2014. “Neural Machine Translation by Jointly Learning to Align and Translate.” arXiv preprint arXiv:1409.0473
8.
BammanDavidO’ConnorBrendanSmithNoah. 2012. “Censorship and Deletion Practices in Chinese Social Media.” First Monday17.
9.
BarberáPablo. 2015. “Birds of the Same Feather Tweet Together: Bayesian Ideal Point Estimation Using Twitter Data.” Political Analysis23:76–91.
10.
BengioYoshuaGoodfellowIan J.CourvilleAaron. 2015. “Deep Learning.” MIT Press book in preparation. Draft chapters available at http://www.iro.umontreal.ca/bengioy/dlbook.
11.
BennettW. LanceSegerbergAlexandra. 2012. “The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics.” Information, Communication & Society15:739–68.
12.
BlecherMarc J.2002. “Hegemony and Workers’ Politics in China.” The China Quarterly170:283–303.
13.
BourgaultAdam. 2015. “Freedom of the Press under Authoritarian Regimes.” Susquehanna University Political Review6:3.
14.
BrunsAxelHighfieldTimBurgessJean. 2013. “The Arab Spring and Social Media Audiences: English and Arabic Twitter Users and Their Networks.” American Behavioral Scientist57:871–98.
15.
BudakCerenWattsDuncan J.2015. “Dissecting the Spirit of Gezi: Influence vs. Selection in the Occupy Gezi Movement.” Sociological Science2:370–97.
16.
CaiYongshun. 2010. Collective Resistance in China: Why Popular Protests Succeed or Fail. Stanford, CA: Stanford University Press.
17.
CastellsManuel. 2015. Networks of Outrage and Hope: Social Movements in the Internet Age. New York: John Wiley & Sons.
18.
ChenXi. 2011. Social Protest and Contentious Authoritarianism in China. Cambridge, UK: Cambridge University Press.
ConneauAlexisSchwenkHolgerBarraultLoïcLecunYann. 2016. “Very Deep Convolutional Networks for Text Classification.” arXiv:1606.01781 [cs] arXiv: 1606.01781.
21.
CroicuMihaiWeidmannNils B.2015. “Improving the Selection of News Reports for Event Coding Using Ensemble Classification.” Research & Politics2(4). https://doi.org/10.1177/2053168015615596.
22.
DahlGeorge E.YuDongDengLiAceroAlex. 2012. “Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition.” IEEE Transactions on Audio, Speech, and Language Processing20:30–42.
23.
DeibertRonald J.2008. “The Geopolitics of Internet Control: Censorship, Sovereignty, and Cyberspace.” Pp. 339–52 in Routledge Handbook of Internet Politics, edited by ChadwickA.HowardP. N.New York: Routledge.
24.
DengYanhuaYangGuobin. 2013. “Pollution and Protest in China: Environmental Mobilization in Context.” The China Quarterly214:321–36.
25.
DhingraBhuwanLiuHanxiaoSalakhutdinovRuslanCohenWilliam W. 2017. “A Comparative Study of Word Embeddings for Reading Comprehension.” arXiv preprint arXiv:1703.00993.
26.
DiamantNeil J.2010. Embattled Glory: Veterans, Military Families, and the Politics of Patriotism in China, 1949–2007. Lanham, MD: Rowman & Littlefield Publishers.
27.
DiamondLarry. 2010. “Liberation Technology.” Journal of Democracy21:69–83.
28.
DimitrovMartinZhangZhu. 2017. “Patterns of Protest Activity in China.” Working paper.
29.
EarlJenniferKimportKatrina. 2008. “The Targets of Online Protest.” Information, Communication & Society11:449–72.
30.
EarlJenniferKimportKatrina. 2011. Digitally Enabled Social Change: Activism in the Internet Age. Boston, MA: MIT Press.
31.
EarlJenniferMartinAndrewMcCarthyJohn D.SouleSarah A.2004. “The Use of Newspaper Data in the Study of Collective Action.” Annual Review of Sociology30:65–80.
32.
EdmondChris. 2013. “Information Manipulation, Coordination, and Regime Change.” The Review of Economic Studies80:1422–58.
33.
EgorovGeorgySoninKonstantin. 2011. “Dictators and Their Viziers: Endogenizing the Loyalty–Competence Trade-Off.” Journal of the European Economic Association9:903–30.
34.
FerdinandPeter. 2000. “The Internet, Democracy and Democratization.” Democratization7:1–17.
GamsonWilliam A.ModiglianiAndre. 1989. “Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach.” American Journal of Sociology95:1–37.
González-BailónSandraBorge-HolthoeferJavierRiveroAlejandroMorenoYamir. 2011. “The Dynamics of Protest Recruitment through an Online Network.” Scientific Reports1.
39.
GuoXiaolin. 2001. “Land Expropriation and Rural Conflicts in China.” The China Quarterly166:422–39.
40.
HannaAlex. 2017. “MPEDS: Automating the Generation of Protest Event Data.”https://mpeds.github.io/
41.
HeKaimingZhangXiangyuRenShaoqingSunJian. 2016. “Deep Residual Learning for Image Recognition.” Pp. 770–78 in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). https://arxiv.org/abs/1512.03385
HintonGeoffreyDengLiYuDongDahlGeorge E.MohamedAbdel-rahmanJaitlyNavdeepSeniorAndrew, et al. 2012. “Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups.” IEEE Signal Processing Magazine29:82–97.
HurstWilliam. 2004. “Understanding Contentious Collective Action by Chinese Laid-Off Workers: The Importance of Regional Political Economy.” Studies in Comparative International Development39:94–120.
46.
HurstWilliamO’BrienKevin J.2002. “China’s Contentious Pensioners.” China Quarterly170:345–60.
47.
HutterSwen. 2014. “Protest Event Analysis and Its Offspring.” Pp. 33–67 in Methodological Practices in Social Movement Research, edited by PortaD. Della. Oxford, UK: Oxford University Press.
48.
JenkinsJ. CraigEckertCraig M.1986. “Channeling Black Insurgency: Elite Patronage and Professional Social Movement Organizations in the Development of the Black Movement.” American Sociological Review51:812–29.
49.
JenkinsJ. CraigPerrowCharles. 1977. “Insurgency of the Powerless: Farm Worker Movements (1946–1972).” American Sociological Review42:249–68.
50.
KalchbrennerNalGrefenstetteEdwardBlunsomPhil. 2014. “A Convolutional Neural Network for Modeling Sentences.” Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics. https://www.aclweb.org/anthology/P14-1062
51.
KimYoon. 2014. “Convolutional Neural Networks for Sentence Classification.” Pp. 1746–51 in Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. https://www.aclweb.org/anthology/D14-1181
52.
KingGaryPanJenniferRobertsMargaret E.2013. “How Censorship in China Allows Government Criticism but Silences Collective Expression.” American Political Science Review107:326–43.
53.
KingGaryPanJenniferRobertsMargaret E.2014. “Reverse-Engineering Censorship in China: Randomized Experimentation and Participant Observation.” Science345:1–10.
54.
KingmaDiederik P.BaJimmy. 2014. “Adam: A Method for Stochastic Optimization.” arXiv preprint arXiv:1412.6980.
55.
KoopmansRuud. 2004. “Movements and Media: Selection Processes and Evolutionary Dynamics in the Public Sphere.” Theory and Society33:367–91.
56.
KoopmansRuudRuchtDieter. 2002. “Protest Event Analysis.” Pp. 231–59 in Methods of Social Movement Research. Vol. 16, edited by KlandermansB.StaggenborgS.Minneapolis: University of Minnesota Press.
57.
KoopmansRuudStathamPaul. 1999. “Political Claims Analysis: Integrating Protest Event and Political Discourse Approaches.” Mobilization: An International Quarterly4:203–21.
58.
KriesiHanspeter. 1995. New Social Movements in Western Europe: A Comparative Analysis. Minneapolis: University of Minnesota Press.
LeCunY.BoserB.DenkerJ. S.HendersonD.HowardR. E.HubbardW.JackelL. D.1989. “Backpropagation Applied to Handwritten Zip Code Recognition.” Neural Computation1:541–51.
62.
LeeChing Kwan. 2007. Against the Law: Labor Protests in China’s Rustbelt and Sunbelt. Berkeley: University of California Press.
63.
LeeSophie J.LiuHowardWardMichael D.2018. “Lost in Space: Geolocation in Event Data.” Pp. 1–18 in Political Science Research and Methods. https://arxiv.org/abs/1611.04837
64.
LiebmanBenjamin L.2013. “Malpractice Mobs: Medical Dispute Resolution in China.” Columbia Law Review113:181–264.
65.
LorentzenPeter. 2013. “Regularizing Rioting: Permitting Protest in an Authoritarian Regime.” Quarterly Journal of Political Science8:127–58.
66.
McAdamDoug. 1982. Political Process and the Development of Black Insurgency, 1930–1970. Chicago: University of Chicago Press.
67.
McAdamDougSuYang. 2002. “The War at Home: Antiwar Protests and Congressional Voting, 1965 to 1973.” American Sociological Review67:696–721.
68.
McAdamDougTarrowSidneyTillyCharles. 2003. Dynamics of Contention. Cambridge, UK: Cambridge University Press.
69.
McCarthyJohn D.McPhailClarkSmithJackie. 1996. “Images of Protest: Dimensions of Selection Bias in Media Coverage of Washington Demonstrations, 1982 and 1991.” American Sociological Review61:478–99.
70.
McDonaldTom. 2016. Social Media in Rural China. London: UCL Press.
71.
McMillanJohnZoidoPablo. 2004. “How to Subvert Democracy: Montesinos in Peru.” Journal of Economic Perspectives18:69–92.
72.
MikolovTomasChenKaiCorradoGregDeanJeffrey. 2013. “Efficient Estimation of Word Representations in Vector Space.” arXiv preprint arXiv:1301.3781.
73.
MikolovTomášKarafiátMartinBurgetLukášČernockýJanKhudanpur.Sanjeev2010. “Recurrent Neural Network Based Language Model.”Eleventh Annual Conference of the International Speech Communication Association.
74.
MillerSeumasSelgelidMichael J.2007. “Ethical and Philosophical Consideration of the Dual-Use Dilemma in the Biological Sciences.” Science and Engineering Ethics13:523–80.
75.
MisloveAlanLehmannSuneAhnYong-YeolOnnelaJukka-PekkaRosenquistJ. Niels. 2011. “Understanding the Demographics of Twitter Users.” Pp. 554–57 in Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media. Vol. 11. Menlo Park, CA: AAAI Press.
76.
NamTaehyun. 2006. “What You Use Matters: Coding Protest Data.” PS: Political Science & Politics39:281–87.
O’BrienKevin J.LiLianjiang. 2006. Rightful Resistance in Rural China. Cambridge, UK: Cambridge University Press.
79.
OliverPamela E.ManeyGregory M.2000. “Political Processes and Local Newspaper Coverage of Protest Events: From Selection Bias to Triadic Interactions.” American Journal of Sociology106:463–505.
80.
OlzakSusan. 1989. “Analysis of Events in the Study of Collective Action.” Annual Review of Sociology15:119–41.
81.
OrtizDavidMyersDanielWallsEugeneDiazMaria-Elena. 2005. “Where Do We Stand with Newspaper Data?”Mobilization: An International Quarterly10:397–419.
82.
PanJennifer. 2017. “How Market Dynamics of Domestic and Foreign Social Media Firms Shape Strategies of Internet Censorship.” Problems of Post-Communism64:167–88.
83.
PanJenniferSiegelAlexandra. 2018. “Physical Repression and Online Dissent in the Saudi Twittersphere.” Working paper.
84.
PascanuRazvanMikolovTomasBengioYoshua. 2013. “On the Difficulty of Training Recurrent Neural Networks.” Pp. 1310–18 in International Conference on Machine Learning. http://proceedings.mlr.press/v28/pascanu13.pdf
85.
PerryElizabeth. 2008. “Permanent Revolution? Continuities and Discontinuities in Chinese Protest.” Pp. 205–16 in Popular Protest in China, edited by O’BrienK.Cambridge, MA: Harvard University Press.
86.
PetersMatthew E.NeumannMarkIyyerMohitGardnerMattClarkChristopherLeeKentonZettlemoyerLuke. 2018. “Deep Contextualized Word Representations.” Proceedings of the North American Chapter of the Association for Computational Linguistics. https://arxiv.org/abs/1802.05365
87.
QinBeiStrömbergDavidWuYanhui. 2017. “Why Does China Allow Freer Social Media? Protests Versus Surveillance and Propaganda.” Journal of Economic Perspectives31:117–40.
88.
QinBeiStrömbergDavidWuYanhui. 2018. “Media Bias in China.” American Economic Review108(9):2442–76.
89.
RainieLeeSmithAaronSchlozmanKay LehmanBradyHenryVerba.Sidney2012. “Social Media and Political Engagement.” Pew Internet and American Life Project.
90.
RamakrishnanNarenButlerPatrickMuthiahSathappanSelfNathanKhandpurRupinderSarafParangWangWei, et al. 2014. “‘Beating the News’ with EMBERS: Forecasting Civil Unrest Using Open Source Indicators.” Pp. 1799–808 in Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ‘14. New York, NY. ACM.
91.
RattaniA.DerakhshaniR.2017. “On Fine-Tuning Convolutional Neural Networks for Smartphone Based Ocular Recognition.” Pp. 762–67 in 2017 IEEE International Joint Conference on Biometrics (IJCB). https://ieeexplore.ieee.org/document/8272767
92.
RenShaoqingHeKaimingGirshickRossSunJian. 2015. “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.” Pp. 91–99 in Advances in Neural Information Processing Systems. https://arxiv.org/abs/1506.01497
93.
RobertsMargaret E.2018. Censored: Distraction and Diversion inside China’s Great Firewall. Princeton, NJ: Princeton University Press.
94.
RomeroDaniel M.MeederBrendanKleinbergJon. 2011. “Differences in the Mechanics of Information Diffusion across Topics: Idioms, Political Hashtags, and Complex Contagion on Twitter.” Pp. 695–704 in Proceedings of the 20th International Conference on World Wide Web.
95.
RuchtDieterKoopmansRuudNeidhardtFriedhelm. 1999. Acts of Dissent: New Developments in the Study of Protest. Lanham, MD: Rowman & Littlefield.
SantosCicero D.ZadroznyBianca. 2014. “Learning Character-Level Representations for Part-of-Speech Tagging.” Pp. 1818–26 in Proceedings of the 31st International Conference on Machine Learning (ICML-14). http://proceedings.mlr.press/v32/santos14.pdf
99.
SarafParangRamakrishnanNaren. 2016. “EMBERS AutoGSR: Automated Coding of Civil Unrest Events.” Pp. 599–608 in Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16. New York: ACM.
100.
SchusterM.PaliwalK. K.1997. “Bidirectional Recurrent Neural Networks.” IEEE Transactions on Signal Processing45:2673–81.
101.
SelgelidMichael J.2013. “Dual-Use Research.” In International Encyclopedia of Ethics. Washington, DC: American Cancer Society.
ShinHoo-ChangRobertsKirkLuLeDemner-FushmanDinaYaoJianhuaSummersRonald M.2016. “Learning to Read Chest X-Rays: Recurrent Neural Cascade Model for Automated Image Annotation.” Pp. 2497–506 in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. https://arxiv.org/abs/1603.08486
104.
SimonyanKarenZisserman.Andrew2015. “Very Deep Convolutional Networks for Large-Scale Image Recognition.” In Proceedings of the Third International Conference on Learning Representations. https://arxiv.org/abs/1409.1556
105.
SmithAaron. 2013. Civic Engagement in the Digital Age. Washington, DC: Pew Research Center.
106.
Steinert-ThrelkeldZachary C.2017. “Spontaneous Collective Action: Peripheral Mobilization during the Arab Spring.” American Political Science Review111:379–403.
107.
Steinert-ThrelkeldZachary C.2018. Twitter as Data. Cambridge, UK: Cambridge University Press.
108.
Steinert-ThrelkeldZachary C.MocanuDeliaVespignaniAlessandroFowlerJames. 2015. “Online Social Networks and Offline Protest.” EPJ Data Science4:19.
109.
SternRachel E.HassidJonathan. 2012. “Amplifying Silence: Uncertainty and Control Parables in Contemporary China.” Comparative Political Studies45:1230–54.
110.
StockmannDaniela. 2013. Media Commercialization and Authoritarian Rule in China. New York: Cambridge University Press.
111.
SuYangHeXin. 2010. “Street as Courtroom: State Accommodation of Labor Protest in South China.” Law & Society Review44:157–84.
SzegedyChristianLiuWeiJiaYangqingSermanetPierreReedScottAnguelovDragomirErhanDumitruVanhouckeVincentRabinovichAndrew. 2015. “Going Deeper with Convolutions.”https://arxiv.org/abs/1409.4842.
114.
TanChade-MengWangYuan-FangLeeChan-Do. 2002. “The Use of Bigrams to Enhance Text Categorization.” Information Processing & Management38:529–46.
115.
TarrowSidney. 2005. The New Transnational Activism. Cambridge, UK: Cambridge University Press.
116.
TheocharisYannisLoweWillVan DethJan W.García-AlbaceteGema. 2015. “Using Twitter to Mobilize Protest Action: Online Mobilization Patterns and Action Repertoires in the Occupy Wall Street, Indignados, and Aganaktismenoi Movements.” Information, Communication & Society18:202–20.
117.
TillyCharles. 2008. Contentious Performances. Cambridge, UK: Cambridge University Press.
118.
TongYanqiLeiShaohua. 2010. “Large-Scale Mass Incidents and Government Responses in China.” International Journal of China Studies1:487–508.
119.
TorresMichelle. 2018. “Give Me the Full Picture: Using Computer Vision to Understand Visual Frames and Political Communication.” Working paper.
120.
TrenthamBarrySokoloffSandraTsangAmieNeysmithSheila. 2015. “Social Media and Senior Citizen Advocacy: An Inclusive Tool to Resist Ageism?”Politics, Groups, and Identities3:558–71.
121.
WangWeiKennedyRyanLazerDavidRamakrishnanNaren. 2016. “Growing Pains for Global Monitoring of Societal Events.” Science353:1502–03.
122.
WangXingyouJiangWeijieLuoZhiyong. 2016. “Combination of Convolutional and Recurrent Neural Network for Sentiment Analysis of Short Texts.” Pp. 2428–37 in Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers. https://www.aclweb.org/anthology/C16-1229
123.
WeissJessica Chen. 2014. Powerful Patriots: Nationalist Protest in China’s Foreign Relations. Oxford, UK: Oxford University Press.
124.
WilsonAshia C.RoelofsRebeccaSternMitchellSrebroNatiRechtBenjamin. 2017. “The Marginal Value of Adaptive Gradient Methods in Machine Learning.” Pp. 4148–58 in Advances in Neural Information Processing Systems. https://arxiv.org/abs/1705.08292
125.
WonDonghyeonSteinert-ThrelkeldZachary C.JooJungseock. 2017. “Protest Activity Detection and Perceived Violence Estimation from Social Media Images.” Pp. 786–94 in Proceedings of the 2017 ACM on Multimedia Conference. New York: ACM.
126.
XiaoYijunChoKyunghyun. 2016. “Efficient Character-Level Document Classification by Combining Convolution and Recurrent Layers.” arXiv:1602.00367 [cs] arXiv: 1602.00367.
127.
YangGuobin. 2003. “The Internet and the Rise of a Transnational Chinese Cultural Sphere.” Media, Culture & Society25:469–90.
128.
ZhangShengZhangXinWangHuiChengJiajunLiPeiDingZhaoyun. 2017. “Chinese Medical Question Answer Matching Using End-to-End Character-Level Multi-Scale CNNs.” Applied Sciences7:767.
129.
ZhouChuntingSunChonglinLiuZhiyuanLauFrancis C. M.2015. “A C-LSTM Neural Network for Text Classification.” CoRR arXiv: 1511.08630.
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
