Multi-Source Assessment of State Stability

The wave of revolutions in the Arab world, commonly referred to as the Arab Spring, took the world by surprise. To some extent the September 2012 consulate and embassy attacks were also unforeseen. Despite the rich literature on interstate conflict, state stability, revolution and regime change these events could not be predicted nor fully accounted for by the existing theoretical traditions in the social sciences. As protests and demonstrations broke out in one country after another, questions arose as to what mechanisms supported the diffusion of ideas and actions that promoted or inhibited violence, and eventually enabled successful regime change. New communication technologies and social media were touted as critical to these revolutions. Traditional media also gave voice to the public concerns and provided information critical to these revolutions. However, the relation between social media, traditional media and social change, the way actors can use those media to promote their agendas, and the impact of the communications in those media on the influence of those actors is not well understood. The proposed research will address this gap. We propose to develop new capabilities in predictive state stability modeling and improve our understanding of the fundamental issues surrounding state stability in a cyber-mediated environment. Our primary objective is to understand the way in which media social and traditional can be used to effect state stability or instability by individuals, groups and corporations. A secondary objective is to identify indicators in social media and traditional media that can be used as signals of changes in trust, norms, lines of stability, and lines of alliance or competition that could predict state instability. It is expected that this research will lead to a new operationally relevant predictive tool for assessing the impact of individuals, groups and messages in social and traditional media on state instability; operationally relevant metrics for assessing trust, state stability, and alliance and changes in these factors; and will support the development and testing of a new media informed theory of state stability. The proposed research addresses three central concerns relative to the relation of media and state stability: First, indicators of state stability/instability, trust, alliance and competition are identified and then the related metrics which can be applied to social and traditional media are developed and tested at the region, state and province level. Second, using statistical, network and visual analytics extracted geo-temporally tagged actor-topic networks and these metrics, are analyzed to assess and predict social behavior at the region, state and province level. Specific attention is paid to assessing who are the trusted information brokers, who are the vulnerable actors, what are the lines of balance, and characterizing the role of social versus traditional media in forging/breaking trust and forging/breaking lines of balance. Third, based on these findings we then turn to asking how can these media be used to influence or inhibit change and so foster changes in state stability, trust, alliance and competition. Using mixed-source data (twitter, blogs, news, trade, geographic information, and archival ethnographic sources) a series of meta-networks linking people, groups, issues, activities, and locations are constructed, and metrics for assessing changes in trust, state stability, and alliance are extracted for a set of countries in the Middle East, Africa and the Pacific Rim. Analysis is done at the region, state and province level thus allowing for improved validation, multi-cultural assessment and a more nuanced understanding of change. A mixed-methods, multimodeling approach is used to support theory development, testing, and model validation. Blending qualitative and quantitative techniques supports the automated coding of media data using text-mining techniques, in-depth analysis of outliers and overall interpretation utilizing culturally informed qualitative ethnographic assessments, and theory testing using traditional and new big-data statistical, network and visual analytics. This research lays the groundwork for a state stability modeling system that is reusable, easily instantiable from empirical open source data, and adaptable to different socio-cultural environments. The techniques and findings pioneered in this work will support all-source data-collection and analysis efforts engaged in for operational needs. The metrics for key indicators, social-topic network models and associated tools developed in this project will provide the DoD with a core operational capability to enhance predictive modeling for regime change, assess social and traditional media data, and assess changes in trust, stability and alliance. The tools and methods developed here, particularly those for capturing, visualizing, analyzing, and fusing information from social media and traditional media are of immediate value to joint HA/DR operations, public-relations operations, irregular warfare operations, intelligence efforts, and IO operations. ASU will collaborate with CMU in support of this effort. ASU will engage in the following activities. 1. Assess original ArabSpring data for lessons learned and provide assessment to team. Collect new data as needed, assess said data in support of addressing questions on trust and state stability, create ORA files for use by research team, and share data and assessment with project team. 2. Extend TweetTracker to capture and record information from business tweets, urls/images on tweets, metadata. Build a comparable system for blogs. Augment both systems so they are interoperable with ORA. Link directly into the ORA visualizer. Explore procedures for meta-data extraction. Develop replay capability. 3. Collaborate with CMU in creating a fused data procedure and database design that supports rapid bigdata processing for the multiple sources of data Twitter, News, Blogs. 4. Test the Tweet/Blogs/REA-ORA cycle and explore using ORA identification of key topics, actors and locations as filters for refined data collection. 5. Develop and test new validation procedures for models developed jointly by the project team. 6. Attend kickoff and annual meetings. 7. Send graduate students to the CASOS Summer Institute. 8. Support CMU in doing demos at the annual meetings and other events. 9. Provide material for the monthly progress reports. This will take the form of bulleted lists sent to Dr. Kathleen Carley, publication/presentation information, and a copy of newly presented publications and presentations. 10. Provide material for the annual progress review and support CMU in development of review material. This will include, but may not be limited to, powerpoint slides, cumulative publication/presentation material, personnel statistics, and requested sections of the overall project annual tech report. 11. Collaborate with CMU in producing project deliverables including publications.