SoCS: Social Media Enhanced Organizational Sensemaking in Emergency Response

Online social networks and always-connected mobile devices have created an immense opportunity that empowers citizens and organizations to communicate and coordinate effectively in the wake of critical events. Specifically, there have been many isolated examples of using Twitter to provide timely and situational information about emergencies to relief organizations, and to conduct ad-hoc coordination. However, there are few attempts that try to understand the full ramifications of using social networks in a more concerted manner for effective organizational sensemaking in such contexts. This multi-disciplinary project, spanning computational and social sciences, seeks to fill this gap.

This project seeks to leverage Twitter posts (tweets) as the primary source of citizen inputs and couple relevant content and network information along with microworld simulations involving human role players to measure effectiveness of various organized sensemaking strategies. To arrive at meaningful summaries of citizen input, tweet content is analyzed using a semantic content analysis by combining natural language techniques that are suitably fused with existing knowledge bases (GeoNames, Wikipedia). Content analysis is further enhanced by innovatively combining it with the dynamic analysis of the twitter network to realize concise and trustworthy information nuggets of potential interest to organizations and citizens. The resulting summaries will be fed to a suitably designed microworld simulation involving human actors to derive realistic settings for modeling disaster situations and typical organizational structures.

This project is expected to have a significant impact in the specific context of disaster and emergency response. However, elements of this research are expected to have much wider utility, for example in the domains of e-commerce, and social reform. From a computational perspective, this project introduces the novel paradigm of people-content-network analysis whose application is not just limited to organized sensemaking. For social scientists, it provides a platform that can be used to assess relative efficacy of various organizational structures using microworld simulations and is expected to provide new insights into the types of social network structures (mix of symmetric and asymmetric) that might be better suitable to propagate information in emergent situations. From an educational standpoint, the majority of funds will be used to train the next generation of interdisciplinary researchers drawn from the computational and social sciences. Participation of underrepresented groups, especially women, will be encouraged, and is anticipated. Datasets and software developed as part of this project will be made available to the broader research community here.

Keywords: Social Networking, Emergency Response, Content Analysis, Network Analysis, Organizational Sensemaking, Collaborative Decision Making, Seeking-Supplying Intent Mining, User Engagement, Coordination.

Collaborative team of Wright State University (WSU) and Ohio State University (OSU):

  • WSU PI/PM: Prof. Amit Sheth.
  • WSU Co-PIs: Prof. Valerie Shalin, Prof. John Flach (Department of Psychology, Human Factors/Industrial Organization Graduate Program).
  • OSU PI: Prof. Srinivasan Parthasarathy.
  • Students: Andrew Hampton, Hemant Purohit, Lu Chen, Shreyansh Bhatt (WSU); Dave Fuhry, Yiye Ruan (OSU) .
  • Funding: This collaborative research is funded by the National Science Foundation under award IIS-1111182 to Wright State University (PI: Amit Sheth) and award IIS-1111118 to Ohio State University (PI: Srinivasan Parthasarathy), 09/01/2011 - 08/31/2014.

Collaborations under the Local and Global Outreach:

Societal Outreach via Media Coverage:

About capability of Twitris research platform to analyze variety of evolving events using similar underlying technologies:

Initiatives of Twitris team: (Refer figures at the end)

Research Tools and Ontology:

  • Twitris: A 360 degree social media analytics platform to assist decision making by providing multi-faceted analyses of social data: Spatio-Temporal-Thematic, People-Content-Network, Sentiment-Emotion-Subjectivity etc.
  • Crisis Computing API interface for providing 'Classification as a Service' on research of seeking-supplying intent classifiers to assist coordination: donation related message, request to help, offer to help, etc. (Also integrated with Ushahidi's CrisisNET project).
  • SoCS Ontology for Crisis Coordination (SOCC): We extend the concepts of domain knowledge­-driven models, MOAC- Management Of A Crisis ontology (Limbu 2012), and UNOCHA's HXL- Humanitarian Exchange Language (Keßler et al. 2013) ontology, with required but missing concepts for organizing data during crisis response coordination for seeker and supplier behavior, and indicators of resource needs using a lexicon. For example, the 'shelter' class contains words 'emergency center,' 'tent,' and 'shelter,' along with lexical alternatives. For the present demonstration, we focus on three resource categories: food, shelter and medical needs. Thus, we endeavor to exploit a minimum, but always expandable subset that provides the maximum coverage while controlling false alarms. For creating lexicons of indicator words for concepts, we relied on various documents collected via interactions with domain experts (Flach et al. 2013), our Community Emergency Response Team (CERT) training, Rural Domestic Preparedness Consortium training, and publically available references (Homeland Security 2010; FEMA 2012; OCHA,Verity 2011). Using a first aid handbook (Swienton and Subbarao 2012), we created an extensive 'medical' subset of emergency indicators, where we identified words which pertained specifically to first aid or injuries and included those words along with variations in tense (i.e., breath, breathing, breathes) and common abbreviations (i.e. mouth to mouth, mouth 2 mouth, CPR). A local expert with FEMA experience augmented the model with additional indicators and provided anecdotal context. The current model with food, medical, and shelter resource indicators contain 43 concepts and 45 relationships. We created this domain model in the OWL language using the Protégé ontology editor (Protégé 2013). Each type of disaster is listed as an entity type with indicators for that disaster listed as individuals under a corresponding indicator entity. Therefore a relationship is declared stating that a particular disaster concept, say Flood, relates by property 'has_a_positive_indicator', with 'Flood_i' indicator entity, that includes all words under that heading. Each disaster has a declared negative relationship with the negative indicator list (e.g., 'erotic' under sexual words indicators) under the entity name Negative_Indicator_i. Finally resources are declared as individuals under the appropriate entity in the same way, but relationships are not explicitly stated with any disaster in order to provide flexibility. [Read more: Purohit et al., JCSCW 2014]
    • Available at: SoCS ontology.
    • Developers: Hemant Purohit, Drew Hampton, Shreyansh Bhatt, Prof. Valerie Shalin. Guidance: Prof. Amit Sheth; External collaborators: Dr. Carlos Castillo (QCRI), Oshani Seveniratne (CSAIL, MIT).

Talks and tutorials:



    Modeling of User Behavior, Location, and Personalization .
  • Wenbo Wang, Lu Chen, Krishnaprasad Thirunarayan and Amit P. Sheth. Cursing in English on Twitter. In ACM Conference on Computer Supported Cooperative Work and Social Computing (CSCW'14), 2014 .
  • Wenbo Wang, Lei Duan, Anirudh Koul, Amit P. Sheth. YouRank: Let User Engagement Rank Microblog Search Results. In the Eighth International AAAI Conference on Weblogs and Social Media (ICWSM'14) 2014.
  • Kapanipathi, P., Jain, P., Venkataramani, C., & Sheth, A. (2014). User Interests Identification on Twitter Using a Hierarchical Knowledge Base. In The Semantic Web: Trends and Challenges (pp. 99-113). Springer International Publishing. 2014.
  • Pavan Kapanipathi, Prateek Jain, Chitra Venkataramani, and Amit Sheth. 2014. Hierarchical interest graph from tweets. In Proceedings of the companion publication of the 23rd international conference on World wide web companion (WWW Companion '14). International World Wide Web Conferences Steering Committee, Republic and Canton of Geneva, Switzerland, 311-312. DOI=10.1145/2567948.2577353
  • Krishnamurthy, R., Kapanipathi, P., Sheth, A. P., & Thirunarayan, K. (2015). Knowledge Enabled Approach to Predict the Location of Twitter Users. In The Semantic Web. Latest Advances and New Domains (pp. 187-201). Springer International Publishing.
    Dynamics of People, Content, and Networks in the Online Communities.
  • Y. Ruan, D. Fuhry, J. Liang, Y. Wang, S. Parthasarathy. (2015) Community Discovery: Simple and Scalable Approaches. In G. Paliouras et al. (Eds.). User Community Discovery. Springer. (In press)
  • Ruan, Y. (2015). Joint Dynamic Online Social Network Analytics Using Network, Content and User Characteristics. Doctoral dissertation, The Ohio State University.
  • H. Purohit, Y. Ruan, D. Fuhry, S. Parthasarathy, A. Sheth. On Understanding Divergence of Online Social Group Discussion. ICWSM 2014.
  • Y. Ruan, S. Parthasarathy. Simultaneous Detection of Communities and Roles from Large Networks. ACM Conference on Online Social Networks (COSN) 2014.
  • Y. Shih, S. Kim, Y. Ruan, J. Cheng, A. Gattani, T. Shi, S. Parthasarathy. Component Detection in Directed Networks. ACM International Conference on Information and Knowledge Management (CIKM) 2014.
  • Y. Ruan, D. Fuhry, S. Parthasarathy. Efficient community detection in large networks using content and links. ACM WWW International Conference on World Wide Web, 2013.
  • S. Parthasarathy. A Scalable Framework for Content+Network Analytics. NSF SoCS Symposium, 2012.
  • Y. Ruan, H. Purohit, D. Fuhry, S. Parthasarthy, A. Sheth. Prediction of Topic Volume on Twitter. 4th Int'l ACM Conference on Web Science (WebSci), 2012.
  • Y. Ruan. On the Interplay of Social Network Connection, User and Content. Doctoral Consortium at NSF SoCS Symposium, 2012.
    H. Purohit, J. Ajmera, S. Joshi, A. Verma, A. Sheth. Finding Influential authors in Brand-page Communities. In Proceedings of the 6th Int'l AAAI Conference on Weblogs and Social Media (ICWSM), 2012.
  • V. Satuluri, S. Parthasarathy, Y. Ruan. Local Graph Sparsification for Scalable Clustering. ACM SIGMOD Int’l Conference on Management of Data, 2011. (Related work done for foundation to build for SoCS work).
    High Dimensional Data analytics.
  • Krishnaprasad Thirunarayan and Amit Sheth. Semantics-empowered Big Data Processing with Applications, In: AI Magazine, 2014, 12 pages.
  • Amit Sheth, Pramod Anantharam, and Krishnaprasad Thirunarayan. Applications of Multimodal Physical (IoT), Cyber and Social Data for Reliable and Actionable Insights. 2nd IEEE C-IOT workshop. 2014. (To appear).
  • D. Fuhry, Y. Zhang, V. Satuluri, A. Nandi, and S. Parathasarathy. PLASMA-HD: Probing the LAttice Structure and MAkeup of High-dimensional Data. (demo) VLDB 2013.
  • X. Yang, A. Ghoting, Y. Ruan, S. Parthasarathy. A Framework for Summarizing and Analyzing Twitter Feeds. ACM SIGKDD Int’l Conference on Knowledge Discovery and Data Mining, 2012.

Check out the summary of our research progress and proposed sensemaking framework here.

Figure 1. Example of coordination during Oklahoma tornado using social media based on automatically identifying and matching needs (on the left) and offers (on the right). The resulting technology of need-offer identification is integrated with the crisis mapping pioneer Ushahidi’s CrisisNET, considered to be the firehose of crisis data, and will be used in managing many future crisis worldwide.


Figure 2. Some of the capabilities of Twitris for emergency management and coordination: (1) shows popular topics – also called social signals (weighted n-grams) related to the chosen event for today and any day of the past since the event was tracked; (2) shows key topics of discussions by locations/regions – states, country: (2.a) to select location of interest—each pin shows a collection of social signals emanating from a location, (2.b) shows popular topic from a location; (3) shows the topical user interaction networks (3.b), and influencers (3.a) with demographics, e.g., with knowledge of user profession (3.c); (4) selects date of analysis, (5) displays tweets (5.a), recent news (5.b), and Wikipedia pages (5.c) related to selected events and social signals; (6) shows event specific multimedia (images (6.a) and videos (6.b)). A number of major natural disasters (tornados, floods, hurricanes/cyclones, earthquakes) were monitored; examples can be found online at A number of additional capabilities related to sentiment, emotion, subjectivity/intent are not shown in this picture but can be seen


Figure 3. Google Crisis Map for cyclone Phalin in Oct 2013, which used crowdsourced data from global digital humanitarian volunteers (aka crisismappers) spearheaded by the SoCS researchers in Twitris team at Kno.e.sis. We thank our collaborators at the Google Crisis Response team and volunteers globally from different organizations (SBTF, OpenCrisis, Info4Disasters, Humanity Road, etc.) and universities, for the tireless effort in this initiative.

We took a similar initiative of crisis mapping during Uttarakhand flash floods in Jun 2013, in collaboration with the Google Crisis Response team and volunteers from universities and prestigious organizations globally (SBTF, Info4, OpenCrisis, HOT, CrisisMappersUK), including a surge support from the Humanity Road. The Hindu newspaper cited this effort: Using crisis mapping to aid Uttarakhand, and for creating awareness about tech-assisted response: Are we missing out on tech-aided disaster management in Uttarakhand?, Jun 2013.

Figure 4. Rescue and Evacuation Stream Map during the historic Jammu & Kashmir Floods in September 2014. Twitris supported the scalable relief effort of initiative, which was cited in several mainstream media, such as Hindustan Times.

Figure 5. Snapshot of Twitris platform based simulation tool for filtering the social stream during functional exercise of emergency response teams in Dayton on the May 28, 2014. Data used for this simulation was based on repurposing of 2013 Boston Bombing dataset given the focus on man-made disasters with Urban focus. The tool provides filtering via search (top left), evolving topics (left pane), and by location (middle pane) for the intractable real-time stream (right pane). The local disaster management and response officials found this to be a highly valuable tool for training/exercise and planning.

Foundation of SOCS:

Check out summaries for related research and projects here.
Student Contact: Hemant Purohit.

A quick summary of our analysis frameworks and systems: (see in full page).

Thesis Defense Resources: