Automated Detection of Terrorist Activities through Link Discovery within Massive Datasets

AI Technologies for Homeland Security

Volume

Papers from the 2005 AAAI Spring Symposium

Contents

This paper describes link discovery technology that is designed to detect threat activities by extracting and piecing together transactional evidence from massive datasets that are composed mostly of noise and clutter. The approach is an integration of several innovative component technologies, including partial pattern matching, hypothesis evaluation and hypothesis merging.