Project: Detecting and Monitoring Terror-related Activities on the Web

While rapid advances in computer and information technologies have made our life easier, they have also caused important opportunities for terror-related organizations to perform their activities. Terrorist organizations, which aim at damaging social, cultural and economical life of the targeted nations, can get valuable benefits from every possible facets of internet and technology by gathering and spreading information in order to promote their goals. Internet has eventually evolved to an uncontrolled environment where it provides a borderless and international medium for message exchange. Most of the active terrorist groups and extremist groups have established their presence on the Internet via Web sites or online bulletin boards. Governments and intelligence agencies are trying to identify terrorist activities on the Web in order to prevent future acts of terror. However, there are several problems preventing effective and efficient discovery of terror-related activities due to an enormous amount of information overload, variety of heterogeneous data sources, non-persistence of information and multi-lingual content. Under this circumstances, there is an urgent need for new technologies and products to assist counter-terrorism experts around the globe in this cyber intelligence effort. _x000D_ _x000D_ In Sakuwassa project, we propose an innovative approach to design and develop a system for detecting and monitoring terror-related activities on the web. The system will be built upon a discriminative and adaptable framework that integrates traditional information retrieval methods on a new architecture. The CO goal of the project is to come up with a working prototype that acts as a web spider in order to detect terror-related content in the web and alert relevant users about ongoing activities. The system is expected to systematically collect and analyze all kinds of terrorist-generated data on the Web. The system architecture will be based on the integration of textual and non-textual information, such as HTML tags, neighbor pages, and image/video content, over two complementary classification methods. It will be designed such that it can support multiple languages and be adapted to other languages. Continuous improvement of the system with expert feedback on collected data/results will be ensured by semi-automated tools. The proposed framework will be adaptable to user demands and needs in different countries._x000D_ _x000D_ The Sakuwassa will target governments, homeland security departments, intelligence agencies and other security services which aim at detecting and monitoring the Web for terror-related activities. It is easy to extend the project to tackle not necessarily terror-related but a general intelligence and security market including big companies, variety of public and private organizations and social groups which may need to detect possible vulnerabilities and threats to their presence. The approaches and methods developed in the project will contribute to advancing the field of intelligence and security informatics. Such advances will help related stakeholders to perform terrorism research and facilitate international security and peace. Therefore, success in this area of research is expected to have implications significantly beyond the scope of the project._x000D_ _x000D_ The project consortium is formed by research intensive SMEs that have a long experience in data mining, machine learning, computational intelligence and natural language processing research, performed development of software tools and provided consulting to other SMEs on project management and software quality assurance, working in two different European countries: Turkey and Denmark. The consortium will work collaboratively within a well structured network of well recognized expert participants.

Acronym Sakuwassa (Reference Number: 5281)
Duration 01/04/2010 - 31/03/2012
Project Topic Sakuwassa will built an adaptable and multi-lingual framework for detecting and monitoring terror-related activities on the Web using text and other type of content.
Project Results
(after finalisation)
We finalized the Sakuwassa product and Mantis, the Turkish P, is now able to market and sell it._x000D_Dezide produced several important key technologies for the final product. These technologies have also been very important for Dezide's own development forward.
Network Eurostars
Call Eurostars Cut-Off 3

Project partner

Number Name Role Country
3 Dezide ApS Partner Denmark
3 MANTIS Yazilim ve Danismanlik Ltd. Sti. Coordinator Turkey
3 IntuView - Pioneering artificial intuition Ltd Observer Israel