This Small Business Innovation Research (SBIR) Phase II project will provide a visual analytics platform that helps visualize how information spreads on the Web through networks of news outlets and social media users. The supported research will extend the interactive visual analytic platform by incorporating better influence modeling, sophisticated propagation cascade models that consider the semantics of the entities and their changing dynamics through time, and new visual paradigms for clustered and grouped data. The interface will allow the end user to manipulate visual representations of how a single press release, news clip, Tweet, or marketing push triggers activity among journalists, micro-bloggers, etc. Public sector policy makers, communications professionals and researchers can use this platform to uncover paradigms in data dissemination, find new ways to influence information dissemination, better inform their leadership, and root out sources of erroneous information online. The Phase II research focuses on dynamic influence monitoring, development of robust propagation cascading models for different social media sites, and the use of visual analytics to understand multi-granularity information propagators. The three areas of research for Phase II are all complementary methods that attempt to characterize, measure, and understand the ubiquitous process of information spread and the influence of individuals in this process as well as allow the user to interact with the underlying data to enhance public outreach.
This grant will continue development of an interactive platform within which users can see and uncover patterns describing how messages are distributed across networks. The tool will locate key influencers, allowing communicators to see exactly how a message was distributed and ways to expedite message delivery during emergencies. Equally important is the ability of the tool to quickly uncover the source(s) and major purveyors of harmful misinformation on the Web. Data and filters further allow users to assess the size and demographic makeup of the audiences being reached enhancing governments interface with the public providing objective measures of the organization's effectiveness in penetrating traditional, new, and social media outlets. This insight will be used to better inform the organization and enhance public awareness of local, state and federal initiatives. Paired with the broader media analysis platform constructed earlier, the supported research will provide a comprehensive means of monitoring and measuring federal, state, and local municipality organizational performance.
Award Title: SBIR Phase II: Innovative Tools to Visualize Digital Media in Digital Era* Federal Award ID: 1127190 Report Submission Period: 08/01/2011 to 04/25/2014 The advent of digital media is drastically changing the manner in which information is created, shared and consumed. These changes enable governments to communicate with a greater number of people far more rapidly today than any other time in history. This revolution in communication affords exceptional opportunities to prevent and mitigate illnesses and calamities through mass communication, but it also introduces new vulnerabilities to governments and to citizens. Threats posed by the spread of harmful misinformation - either by design or by accident - across new media must be addressed with the advent of new technologies that improve rapid response in identifying and tracking sources and major purveyors of misinformation online. Very often, it is not until a message has already gone viral that an agency learns misinformation is out there. As such, government agencies need tools that would allow them to visualize the pathway of their communications in order to locate key influencers, uncover new ways to expedite information flow, and locate sources of erroneous information that may pose threats to public health and safety. This Small Business Innovation Research Phase II research focused on overcoming the challenges discovered in the NSF SBIR Phase I research including: 1) customizing code to avoid entity duplication and extract relevant information; 2) developing multiple complimentary metrics to identify digital influence; and 3) developing visual analytic tools for interactive analysis of social content. Phase II research also expanded three areas of research including: 1) testing known spread propagation algorithms against social media diffusion patterns, 2) dynamic influence monitoring, and 3) the use of visual analytics to understand multi-granularity information propagators. The supported research provided a robust media monitoring and visual analytics platform that eliminates the mystery behind how information spreads across the Web through networks of news outlets and social media users by incorporating better influence modeling, sophisticated propagation cascade models that consider the semantics of the entities and their changing dynamics through time, and new visual paradigms for clustered and grouped data. The software allows the end user to view and manipulate a variety of visual representations of how a single piece of digital information spreads across the Web, and is capable of visualizing both large and batches of information. Public sector policy makers, communications professionals and researchers can use this platform to uncover patterns in data dissemination, identify message influencers and mavens, find new ways to influence information dissemination, better inform their leadership about the impact of an event based on the number of people reached, track key message uptake across specific groups of users, and quickly trace sources of erroneous information online to its source. The system also identifies patterns of data dissemination, which may be used to identify fake or nefarious social media accounts used by those plotting against the United State government or its laws. *"This Project Summary Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content."