You are here

Situational Awareness System


TECHNOLOGY AREA(S): Information Systems

OBJECTIVE: Develop and deliver a stand-alone system for online threat detection and behavioral analysis for enhanced situational awareness.

DESCRIPTION: With the ever increasing number of social networking platforms and ensuing user-generated soft-data content, military-based intelligence is more than ever in need of a system to automate and streamline its collection, storage, analysis, and visualization capabilities and to incorporate this information into appropriate analyses for improved military intelligence assessment and soldier situational awareness.

Current intelligence gathering and analysis techniques are significantly human-labor driven for identifying, searching, discovering, and analyzing content on various web sites. Worst case, results are inefficient, incomplete, and disorganized, resulting in dubious outcomes, and in the best case are so labor intensive they can be inaccurate because they are based on outdated stale data. Recent technological and methodological developments in data science allow technically advanced users to scrape [1] massive quantities of data with modest effort from sources like Twitter and WordPress [2,3]. Likewise, advancements in processing textual data via social network analysis [4], natural language processing [5], and graph visualization for community detection [6,7] allow skilled users to explore and analyze big data. However, such advanced techniques are only accessible to highly skilled, technically savvy people. To make use of such data for enhanced situational awareness by a greater number of people, we need to bring together various methodologies to develop a comprehensive, scalable, user-friendly system to (A) collect, (B) store, (C) analyze, and (D) visualize public data.

The first challenge (collect) is to create a simplified process, allowing non-technical end-users to select from one or more public sites for data scraping based on a set of input parameters (e.g., name, timeframe, keywords, hashtags) [1]. The second challenge (store) is to create an organizational system flexible enough to store variable datatypes stemming from a plethora of data sources including yet to be created social media platforms. The third challenge (analyze) is to design and leverage existing algorithms for data analysis. Analysis capabilities must include, but are not limited to, social network analysis of single graphs and multi-graphs; flow analysis to assess information cascades throughout the network; individual-level behavior pattern analysis with the capacity to identify behavioral changes, and community detection [6,7]. Additional analysis capabilities should incorporate the latest analytic capabilities based on language analysis such as deception detection [8] and indicators of hierarchical positioning [9]. The fourth challenge (visualize) is to create scalable visualization techniques that will allow the user to explore individual profile information; fluidly visualize single or multimodal network graphs; drill through graphs to uncover the underlying data; and show how individual behavior patterns change over time (e.g., frequency of tweeting; length of blog posts) [6].

The implemented system will have a small form factor that is multiplatform, portable, and scalable. The system will provide the ability to choose multiple algorithms for analysis based on user needs. Additionally, the GUI must be turnkey, with an easy to use interface for non-technical end-users. The system will contain a searchable database with available communications from multiple sources such as Twitter, Facebook, LinkedIn, and blogs, and should allow for multiple data types such as text, pictures, audio, and video. All stored data must retain relevant meta-data like sender, receiver, date, time, and geo-tags. The system should provide efficient analysis capabilities including the ability to create search profiles, custom categorizations – both emergent and pre-defined, identification of behavior patterns (e.g., an individual posts most frequently during late night hours), and identify changes in behavior patterns (e.g., individual suddenly posts in afternoon).

PHASE I: The Phase I effort will address the first two challenges by developing and demonstrating a prototype system capable of running from a portable device with an intuitive user interface that will allow for data scraping from multiple sources based on a single set of input parameters. The prototype solution must be capable of running off a stand-alone USB drive without the need to install files on the host machine; moreover, the software should not connect to central server for data storage or processing (that is, no cloud-based solutions will be accepted). The software tool should be designed in a manner to aggregate disparate data types from a variety of sources and be modular in design so it can be easily updated as social network companies release new or change APIs. The specific data to be scraped should be driven by an informed military need.

PHASE II: The Phase II effort will address the third and fourth challenges by concentrating on the design and development of analytic capabilities to create a composite picture from multiple data sources and providing informative, scalable visualization capabilities for both data and analytics. Additionally, indicate and flag changes in behavior based on communication patterns obtained through different social media inputs. In addition, Phase II will develop valid social network link prediction analytics and community structure analysis tools. The offeror must demonstrate a clear understanding of analytics relevant to military needs.

PHASE III DUAL USE APPLICATIONS: Phase III efforts will be directed toward refining a final deployable design with sophisticated, cross-platform GUI; incorporating design modifications based on results from tests conducted during Phase II; and improving engineering/form factors, equipment hardening, and manufacturability designs to meet U.S. Army Concept of Operations (CONOPS) and end-user requirements.


    • Marres, Noortje and Weltevrede, Esther. 2013. Scraping the Social? Issues in live social research. Journal of Cultural Economy, 6(3), pp. 313-335.


    • Côté, Isabelle. 2013. Fieldwork in the Era of Social Media: Opportunities and Challenges. PS: Political Science & Politics, 46(03), pp 615-619.


    • Boyd, Ellison. 2007. Social Network Sites: Definition, History, and Scholarship. Journal of Computer-Mediated Communication. 13(1), pp. 210-230.


    • Pattuelli, M. C., and Miller, M. 2015. Semantic network edges: a human-machine approach to represent typed relations in social networks. Journal of Knowledge Management, 19(1), pp.71 – 81.


    • Soeken, M. & Drechsler, R. 2014. NLP-Assisted Model Generation. Springer.


    • Nikolaev, A. G., Razib, R., Kucheriya, A. 2015. On efficient use of entropy centrality for social network analysis and community detection. Social Networks. 40, pp. 154-162.


    • Bothorel, C., Au - Cruz, J., Magnani, M., 2015. Clustering attributed graphs: Models, measures and methods. Network Science, available on CJO2015. doi:10.1017/nws.2015.9.


    • Hauch, V., Blandón-Gitlin, I., Masip, J., & Sporer, S. L. 2014. Are Computers Effective Lie Detectors? A Meta-Analysis of Linguistic Cues to Deception. Personality and Social Psychology Review. pp. 1-36.


  • Kacewicz, E., Pennebaker, J. W., Davis, M., Jeon, M., & Graesser, A. C., 2013. Pronoun Use Reflects Standings in Social Hierarchies. Journal of Language and Social Psychology. pp. 1–19


  • TPOC-1: Dr. Edward Palazzolo
  • Phone: 919-549-4234
  • Email:
  • TPOC-2: Dr. Paul Baker
  • Phone: 919-549-4202
  • Email:
US Flag An Official Website of the United States Government