WIT Press


Link Mining For Contextual Prioritization Of Tactical Information

Price

Free (open access)

Paper DOI

10.2495/DATA080251

Volume

40

Pages

10

Page Range

259 - 268

Published

2008

Size

493 kb

Author(s)

R. E. Marmelstein, M. G. Devito, C. Mungo, G. Papanagopoulos, K. Patel & M. H. Linderman

Abstract

To address the problem of information overload in military Command and Control (C2), we have prototyped a Tactical Information Prioritization System (TIPS). TIPS performs link mining on relational, C2 databases in order to dynamically determine the relevance of a given report to a generic military unit of interest. Emerging C2 database schemas, such as C2 Integrated Exchange Data Model (C2IEDM), support this goal by providing multiple, overlapping pathways to relate disparate battlespace objects. By leveraging the C2IEDM schema, we have developed a set of queries to mine context-sensitive relationships for a given report/unit pairing. These findings are then interpreted by a Bayesian Network to prioritize a set of reports for transmission to the specified unit. Our testing shows that the TIPS recommendations approximate those of group of human analysts. Keywords: context awareness, command and control, link mining, Bayesian Networks, information prioritization, C2IEDM. 1 Background Communication is always a difficult task; this is especially true in combat. Warfighters must receive the information necessary for them to accomplish their mission in a timely manner. Conversely, an excess of information can degrade unit effectiveness and jeopardize mission success. Supplying the right

Keywords

context awareness, command and control, link mining, Bayesian Networks, information prioritization, C2IEDM.