The Multi-Source Signatures for Nuclear Programs project, part of Pacific Northwest National Laboratory's (PNNL's) Signature Discovery Initiative, seeks to computationally capture expert assessment of multi-type information to assess nuclear activities through a series of Bayesian network (BN) models. Information types may include text, sensor output, imagery, or audio/video files. The BN models incorporate knowledge from a diverse range of information sources to help assess a country's nuclear activities. The models span engineering topic areas, country-level indicators, and facility-specific characteristics. To illustrate the development, calibration, and use of BN models for multi-source assessment, we present a model that predicts a country's likelihood to participate in the international nuclear nonproliferation regime. We validate this model by examining the extent to which the model helps non-experts arrive at conclusions similar to those provided by nuclear proliferation experts. We also describe the PNNL-developed software used throughout the lifecycle of the Bayesian network model development.

Bayesian networks provide a general framework with which to model many natural phenomena. The mathematical nature of Bayesian networks enables a plethora of model validation and calibration techniques: e.g. parameter learning, structure learning, goodness of fit tests, and diagnostic checking of the model assumptions. However, they are not free of shortcomings. With regard to parameter learning, in practice it is not uncommon to find oneself lacking adequate data to reliably estimate all model parameters. In this paper we present the early development of a novel application of conjoint analysis as a method for eliciting and modeling expert opinions and for using the results in a methodology for calibrating the parameters of a Bayesian network.

Text streams are a fundamental source of information that can be used to detect and characterize strategic intent of individuals and organizations as well as for detecting abrupt or surprising events within communities. In this paper we describe our algorithm development and analysis methodology for mining the evolving content in text streams. Text streams include news, press releases from organizations, speeches, Internet blogs, etc. Specifically, an analyst may need to know if and when the topic within a text stream changes. Much of the current text-feature methodology is focused on understanding and analyzing a single static collection of text documents. Corresponding analytic activities include summarizing the contents of the collection, grouping the documents based on similarity of content, and calculating concise summaries of the resulting groups. The approach reported here focuses on taking advantage of the temporal characteristics in a text stream to identify relevant features (such as change in content), and also on the analysis and algorithmic methodology to communicate these characteristics to a user. We present a variety of algorithms for detecting essential features within a text stream. Our approach for communicating the information back to the user is to identify feature (word/phrase) groups. These resulting algorithms form the basis of developing software tools for a user to analyze and understand the content of text streams. We present analysis results using both news information and abstracts from technical articles, and show how these algorithms provide understanding of the contents of these text streams. A critical finding is that the characteristics we used to identify features in a text stream are uncorrelated with the characteristics used to identify features in a static document collection.

The ability to support creation and parallel analysis of competing scenarios is perhaps the greatest single challenge for today's intelligence analysis systems. Dempster-Shafer theory provides an evidentiary reasoning methodology for scenario construction and analysis that offers potential advantages when compared to other approaches such as Bayesian nets as it places less conceptual load on the analyst by not requiring the complete specification of joint probability distributions. This paper presents a method that can further reduce the conceptual load by taking advantage of hierarchically structured indicators. We present a novel interface for this layered, Dempster-Shafer evidentiary reasoning approach and demonstrate the utility of this interface with reference to analysis problems focusing on comparing distinct hypotheses.

The ThemeRiver visualization depicts thematic variations over time within a large collection of documents. The thematic changes are shown in the context of a time-line and corresponding external events. The focus on temporal thematic change within a context framework allows a user to discern patterns that suggest relationships or trends. For example, the sudden change of thematic strength following an external event may indicate a causal relationship. Such patterns are not readily accessible in other visualizations of the data. We use a river metaphor to convey several key notions. The document collection's time-line, selected thematic content and thematic strength are indicated by the river's directed flow, composition and changing width, respectively. The directed flow from left to right is interpreted as movement through time and the horizontal distance between two points on the river defines a time interval. At any point in time, the vertical distance, or width, of the river indicates the collective strength of the selected themes. Colored "currents" flowing within the river represent individual themes. A current's vertical width narrows or broadens to indicate decreases or increases in the strength of the individual theme.