We describe the design, implementation, and evaluation of EMBERS, an automated, 24x7 continuous system for fore- casting civil unrest across 10 countries of Latin America using open source indicators such as tweets, news sources, blogs, economic indicators, and other data sources. Un- like retrospective studies, EMBERS has been making fore- casts into the future since Nov 2012 which have been (and continue to be) evaluated by an independent T&E team (MITRE). Of note, EMBERS has successfully forecast the June 2013 protests in Brazil and Feb 2014 violent protests in Venezuela. We outline the system architecture of EMBERS, individual models that leverage specific data sources, and a fusion and suppression engine that supports trading off spe- cific evaluation criteria. EMBERS also provides an audit trail interface that enables the investigation of why specific predictions were made along with the data utilized for fore- casting. Through numerous evaluations, we demonstrate the superiority of EMBERS over baserate methods and its capability to forecast significant societal happenings.
%0 Conference Paper
%1 Ramakrishnan:2014:BNE:2623330.2623373
%A Ramakrishnan, Naren
%A Butler, Patrick
%A Muthiah, Sathappan
%A Self, Nathan
%A Khandpur, Rupinder
%A <b>Saraf, Parang</b>
%A Wang, Wei
%A Cadena, Jose
%A Vullikanti, Anil
%A Korkmaz, Gizem
%A Kuhlman, Chris
%A Marathe, Achla
%A Zhao, Liang
%A Hua, Ting
%A Chen, Feng
%A Lu, Chang Tien
%A Huang, Bert
%A Srinivasan, Aravind
%A Trinh, Khoa
%A Getoor, Lise
%A Katz, Graham
%A Doyle, Andy
%A Ackermann, Chris
%A Zavorin, Ilya
%A Ford, Jim
%A Summers, Kristen
%A Fayed, Youssef
%A Arredondo, Jaime
%A Gupta, Dipak
%A Mares, David
%B Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
%C New York, NY, USA
%D 2014
%I ACM
%K civil embers event forecasting indicators myown news open source unrest
%P 1799--1808
%R 10.1145/2623330.2623373
%T 'Beating the News' with EMBERS: Forecasting Civil Unrest Using Open Source Indicators
%U http://doi.acm.org/10.1145/2623330.2623373
%X We describe the design, implementation, and evaluation of EMBERS, an automated, 24x7 continuous system for fore- casting civil unrest across 10 countries of Latin America using open source indicators such as tweets, news sources, blogs, economic indicators, and other data sources. Un- like retrospective studies, EMBERS has been making fore- casts into the future since Nov 2012 which have been (and continue to be) evaluated by an independent T&E team (MITRE). Of note, EMBERS has successfully forecast the June 2013 protests in Brazil and Feb 2014 violent protests in Venezuela. We outline the system architecture of EMBERS, individual models that leverage specific data sources, and a fusion and suppression engine that supports trading off spe- cific evaluation criteria. EMBERS also provides an audit trail interface that enables the investigation of why specific predictions were made along with the data utilized for fore- casting. Through numerous evaluations, we demonstrate the superiority of EMBERS over baserate methods and its capability to forecast significant societal happenings.
%@ 978-1-4503-2956-9
@inproceedings{Ramakrishnan:2014:BNE:2623330.2623373,
abstract = {We describe the design, implementation, and evaluation of EMBERS, an automated, 24x7 continuous system for fore- casting civil unrest across 10 countries of Latin America using open source indicators such as tweets, news sources, blogs, economic indicators, and other data sources. Un- like retrospective studies, EMBERS has been making fore- casts into the future since Nov 2012 which have been (and continue to be) evaluated by an independent T&E team (MITRE). Of note, EMBERS has successfully forecast the June 2013 protests in Brazil and Feb 2014 violent protests in Venezuela. We outline the system architecture of EMBERS, individual models that leverage specific data sources, and a fusion and suppression engine that supports trading off spe- cific evaluation criteria. EMBERS also provides an audit trail interface that enables the investigation of why specific predictions were made along with the data utilized for fore- casting. Through numerous evaluations, we demonstrate the superiority of EMBERS over baserate methods and its capability to forecast significant societal happenings.},
acmid = {2623373},
added-at = {2014-10-23T22:34:27.000+0200},
address = {New York, NY, USA},
author = {Ramakrishnan, Naren and Butler, Patrick and Muthiah, Sathappan and Self, Nathan and Khandpur, Rupinder and <b>Saraf, Parang</b> and Wang, Wei and Cadena, Jose and Vullikanti, Anil and Korkmaz, Gizem and Kuhlman, Chris and Marathe, Achla and Zhao, Liang and Hua, Ting and Chen, Feng and Lu, Chang Tien and Huang, Bert and Srinivasan, Aravind and Trinh, Khoa and Getoor, Lise and Katz, Graham and Doyle, Andy and Ackermann, Chris and Zavorin, Ilya and Ford, Jim and Summers, Kristen and Fayed, Youssef and Arredondo, Jaime and Gupta, Dipak and Mares, David},
biburl = {https://www.bibsonomy.org/bibtex/2aed07162105823c00a96a5ed50b129df/parangsaraf},
booktitle = {Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
doi = {10.1145/2623330.2623373},
interhash = {c113fc0316893b05a27225dd5aadec7b},
intrahash = {aed07162105823c00a96a5ed50b129df},
isbn = {978-1-4503-2956-9},
keywords = {civil embers event forecasting indicators myown news open source unrest},
location = {New York, New York, USA},
numpages = {10},
pages = {1799--1808},
publisher = {ACM},
series = {KDD '14},
timestamp = {2014-10-23T23:20:38.000+0200},
title = {'Beating the News' with EMBERS: Forecasting Civil Unrest Using Open Source Indicators},
url = {http://doi.acm.org/10.1145/2623330.2623373},
year = 2014
}