A CBR Approach to Asymmetric Plan Detection
Daniel Y. Fu, Stottler Henke, Emilio Remolina, Jim Eilbert · 2003
We describe an approach to the problem of detecting the execution of mission plans by the unconventional side in asymmetric warfare. This problem is characterized by actors who go to great lengths to avoid detection, while most of their actions are seemingly innocuous unless placed in a broader context. The problem is to find threatening patterns of action in a data collection characterized as massive, relational, incomplete, noisy, and corrupt. In this paper we describe Sibyl: a subsystem embodying a case-based reasoning approach to automated plan detection. Sibyl features a "spanning case base" that covers the space of theoretical scenarios. It uses each case in a state-space search algorithm by adapting case elements to the data. A simulator for Russian organized crime was used to generate case and test data. We describe Sibyl's algorithm and experimental results used in this approach.