Progress beyond the state of the art
DANTE project aims to deliver the following main results:
- A set of innovative services for the automated detection and analysis of online multimedia and multi-language contents, relying on existing solutions, frameworks, and prototypes owned by the project partners or available as Open source software;
- A TRL-7 system prototype exploiting and integrating the functionalities and capabilities offered by the above-mentioned services: the system will deliver intelligence and collaboration applications as well analytics services for LEAs.
In particular, following end-users needs and LEAs requirements DANTE is focused on enhancing the state of the art and introducing potential innovation in:
1) Visual-based semantic concept detection
2) Audio analysis
3) Textual analysis
4) Knowledge Mining, Fusion and Intelligence Tools
Potential Impacts
The project is implementing the following automated functionalities able to provide LEAs with performing services for supporting, improving and speed up investigative activities:
1) To promptly find, monitor and manage sources of potentially relevant terrorist-related data in the surface Web;
2) To promptly find, monitor and manage sources of potentially relevant terrorist-related data in the Deep Web, including the Dark Nets;
3) To accurately and promptly identify, analyse, assess, and filter potential terrorist-generated and terrorist-related multimedia data and contents in multiple languages;
4) To accurately classify, categorize, and group potential terrorist-related and/or terrorist-generated multimedia data and contents in multiple languages (including to group information from several sources that support same history, to isolate potential sources of “disinformation”, etc.);
5) To store and preserve captured “relevant” data in dedicated and secure data management infrastructures and repositories to enable and support further forensic analysis, respecting the chain of custody;
6) To accurately summarize relevant terrorist-related multimedia data and contents;
7) To promptly detect, identify, analyse, and monitor potential terrorist-related activities, with a special focus on raising funds.
8) To detect and identify potential undiscovered terrorists and terrorist-related individuals (including the mapping of digital identity and pseudonyms with physical identity, with the aim of finding the original author of or people mentioned in terrorist-related contents);
9) To accurately detect and identify potential undiscovered terrorism-related groups of people and online communities (i.e. global terrorist organizations, grassroots terrorist cells);
10) To perform large-scale temporal analysis of terrorism trends, by analysing the collected and extracted terrorism-related relevant events to achieve better understanding of terrorism phenomena (including automatic analysis of dissemination contents such as news, papers, reports, etc.).
Such functionalities, aimed at delivering a novel and breakthrough way of detecting, identifying and monitoring terrorist-related contents and activities on the Web, including the less accessible part.
