Uncertainty-Sensitive Heterogeneous Information Fusion: Assessing Threat with Soft, Uncertain, and Conflicting Evidence

Abstract

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited. Permission is given to duplicate this document for personal use only, as long as it is unaltered and complete. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial use. For information on reprint and linking permissions, please visit www.rand.org/pubs/permissions.html.The RAND Corporation is a research organization that develops solutions to public policy challenges to help make communities throughout the world safer and more secure, healthier and more prosperous. RAND is nonprofit, nonpartisan, and committed to the public interest.RAND's publications do not necessarily reflect the opinions of its research clients and sponsors. For more information on this publication, visit www.rand.org/t/RR1200Published by the RAND Corporation, Santa Monica, Calif. © Copyright 2016 RAND CorporationR® is a registered trademark. Cover Image: Fotolia iii PrefaceThis study was sponsored by the Office of Naval Research (ONR). It was initiated by Ivy Estabrooke and continued under Lee Mastrioanni, thrust manager, and Joong Kim, program officer, in ONR's Expeditionary Maneuver Warfare and Combatting Terrorism Department (Code 30). Comments and questions are welcome and should be addressed to Paul K. Davis at pdavis@rand.org.The report documents a basic research project. It is technical in nature and intended for researchers or managers of technical research potentially interested in information fusion for such domains as counterterrorism, law enforcement, and intelligence. Some of the ideas and methods will be of interest to the larger community of researchers involved with information fusion.This research was sponsored by the Office of Naval Research and conducted within the International Security and Defense Policy Center of the RAND National Defense Research Institute, a federally funded research and development center sponsored by the Office of the Secretary of Defense, the Joint Staff, the Unified Combatant Commands, the Navy, the Marine Corps, the defense agencies, and the defense Intelligence Community.For more information on the International Security and Defense Policy Center, see http://www.rand.org/nsrd/ndri/centers/isdp.html or contact the director (contact information is provided on web page). Summary ObjectivesMilitary and other government organizations put substantial effort into detecting and thwarting attacks such as those by suicide bombers or involving improvised explosive devices. Such attacks may be against military or government installations in the United States or abroad, civilian infrastructure, or any of many other targets. An element of thwarting such attacks is observing suspicious individuals over time with such diverse means as cameras, scanners, and other devices; travel records; behavioral observations; and intelligenc...

Priority Challenges for Social and Behavioral Research and Its Modeling

“…analy sis needs to consider uncertainties in the model itself-i.e., structural uncertainties (Davis et al, 2016)-not just the pa ram e ter values of a given model.…”

“…This poses additional prob lems for modeling platforms if they are to support uncertainty analy sis broadly. Dealing with model uncertainty is a long-standing prob lem, but much can be done, as illustrated in recent work on heterogeneous information fusion (Davis et al, 2016).…”

Representing qualitative social science in computational models to aid reasoning under uncertainty: National security examples

“…In our work we have frequently found it useful to separate discussion of snapshot causality and system dynamics (including feedbacks), to suppress some cross-linkages that are relatively weak, and to define variables of a tree left to right so that the higher level factors can be treated as logically independent. 29 Reality is complex; tricks are required to glimpse various aspects in seemingly simple ways.…”

“…This basic research study investigated ways to fuse heterogeneous information as part of detecting terrorists. 29,51 It sought to combine information types ranging from digital records on an individual’s background (e.g., prior arrests and convictions, travel history, weapons purchases), human sources (e.g., informants), and behavioral observations (e.g., deceptive behavior in an airport line). The information could be fragmentary, ambiguous, contradictory, or even deceptive.…”

Simple Culture-Informed Models of the Adversary

“…In that study we attempted to confront all the uncertainties of the problem, structural, parametric, and procedural. For example, we built in the ability to easily vary the functional forms for how factors combine, the values of related parameters, and even the order of processing reports (Davis et al, 2016). The same methods would apply for cognitive modeling.…”