Open Source Analysis Services

CAMRIS International Solution: Synopsis

Our solution set includes unique technology tools that, when combined with sophisticated analysis, enables the detection, tracking and measurement of content and trends in Open Source Information (OSI). Delivering much more than traditional “news clipping” efforts, OSI mines the world’s unguarded knowledge gleaned from all public channels, anything that contains information, in any language, and from any country - such as broadcast television and radio, printed media such as phone books, newspapers and magazines, and internet discussions in the form of internet publications, web blogs, and chat rooms, and other sources that are “open” (as opposed to “secure” information that is not available in these media sources). OSI also includes “deep web” data sources.

There is a vast volume of information globally; OSI gathering and analysis is one way to help cope with the magnitude of data. The data is archived, and analysts can then draw upon it in response to queries from all levels of government. By capturing and analyzing OSI content, an interested entity – government, research or commercial – can know with the greatest possible certainty what is being discussed by the public and what the tone and sentiment of discussions are. The public may include a society or country at large; or it can transcend national boundaries to include “communities” (such as “youth” or “Europeans” or “European youth”); it can also include identifiable groups within countries, such as “those who get their information from the Internet” or “TV news channels.”

By capturing and analyzing OSI, our solution enables our customers to know more than breaking news, they can understand the underlying sentiment and what opinions are expressed in the open source environment. It also makes it possible to map content and opinions to events. For example, it allows the exploration of whether sentiment expressed in OSI on topics such as the condition of the economy or the validity or usefulness of certain government policies, may be associated with events occurring independently of the open source environment.

CAMRIS currently utilizes subcomponents of the technology to provide foreign media reporting and analysis to government institutions. We are able to deliver targeted, timely and authoritative OSI for analysis, operations and policymaking.

Characteristics and Functions

Our technology solution can capture and store content from many sources and in many languages. Its multilingual data capture system is its core system. We integrate the best available technologies for receiving information from or searching for information in open sources. At this level, it uses a number of best of breed commercial products. Our solution set assembles these combined capabilities into an engine that drives customizable tools for monitoring and analysis. If the capture and storage of OSI content is the core of our solution, then its analytic tools are its nerve center. We draw upon tools to include nonlinear models for monitoring, analysis, and measurement, ultimately for predictions. These tools are built and customized by CAMRIS analysts and scientist.

Our solution is a “learning system” – built on artificial intelligence architecture – with the result that models run by the system continuously capture, search, refresh, test, recalibrate and re-verify. The distinguishing characteristics are therefore:

• Nonlinear modeling – when the system identifies relationships and develops new models they are based on the best available nonlinear mathematical approaches
• Dynamic systems – the system interacts with the open source environment and learns from that environment
• Intercept – the system captures information from the entire open source environment

If required, the system can operate in every major language as well as several local languages and dialects, and forms concepts within the logical structures while also providing transplantations to other languages. The system also stores information to enable trend analysis and that searches back through previous open source information – which may have occurred months or even years ago to uncover patterns and to find other patterns of significance. The system is targeted to monitor and track the state of a “nation” but can just as well monitor and track the “state” of a product or brand.

The system is organized so that the data collected can be analyzed and synthesized by specialists including the development of “alerts” as well as trends. Intelligent tools are provided that act to multiply the analyst ability to monitor and track subject matter. The diagrams below provide a simplified view of how the components of the system operate together.

CAMRIS advanced tool set provide the following functionality:

• Opinion and Discussion Monitoring - As part of its routine monitoring activities, the system is set to monitor web blogs and chat rooms. As a result an expert user can identify patterns within these web-based communication channels and groups’ channels by category of intensity and opinion. These categories guide analysts to more closely focus on certain blogs, chat rooms or other sources. Because the gathered information forms a historical data base, it can be researched when new information might reshape a previous analysis. The system can be further customized to develop and test counter-measures that have a positive impact on observable patterns.
  
• Event Analysis and Anticipation - When an event of consequence occurs the system can be configured to detect unusual chatter in the web communication channels allowing trained users to search for clusters of communication that may be related to this event. Because the system can store archived information, analysts and investigators could use it to scan for patterns leading up to this event.
  
• Visual Recognition – The system includes visual identification tools that provide video frame analysis capabilities both of still images and motion video (including “live” television). Visual recognition can be matched against a library of known images. Additionally, individual occurrences, similarities and statistical relationships across observations, time and locations can significantly deepen understanding of the data being analyzed.
  

Options

The system filters can be set to capture some, most or all of open source content. Reliability of results increases as these filters are set to capture as much content as possible. Costs increase proportionately. A low capture filter will capture the most easily accessible information, such as cable broadcast content available from international commercial carries, or newspapers that publish content on the internet. A medium capture filter will include installations of local capture and store and forward devices, such as over the air broadcasts that may be captured locally and feed to the system by means of local store-and-forward servers, as well as tools such as “blog spiders” to “crawl” into web pages and chat rooms and capture content. A complete capture can include local radio waves, and even local flyers and other media that can be locally scanned and forwarded.

Our is not an “off the shelf” system. The system is customized and deployed for a determined period of time with user defined objectives. Costs of deployment vary with the extent of capture, the number of languages, the extent of customized analytic models that are developed, the locations involved, and other variables.