Machine learning towards a universal situational awareness (USA)

This paper addresses the problem of detecting and countering potential terrorist threats. It describes a synergistic conjunction of intelligent technologies with a view towards that goal. In particular, this paper shows that knowledge acquisition is to be properly viewed as a symbiotic activity of human and machine. The role of the human is to provide the novel or random information and the role of the machine is to reuse it and all of its symmetric derivations. As more and more knowledge is acquired, the number of symmetries that can be related to an arbitrary piece of knowledge necessarily increases. This means that system intelligence, that is, utility, is necessarily a phenomenon of scale given the proposed architecture. A natural language front-end interface is described-which is designed to reduce the impedance mismatch between the human and the machine. Most significantly, the effective translation of natural language semantics is shown to critically depend on an accelerated capability for learning. Next, conversational natural language is mapped onto a set of procedures which effect directed mining and linking operations on a relational database. Finally, projected and restricted database views are analyzed by a distinct high-level knowledge base to generate natural language reports. Most importantly, these reports provide the analyst with immediate feedback that serves to catalyze the generation of subsequent queries.