Mixed-initiative data mining with Bayesian networks

Stärk, Robert F.; Farry, Michael; Pfautz, Jonathan

doi:10.1109/cogsima.2012.6188360

Cited by 3 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Stark et al (2012), it is proposed a mixed approach combining DM and Bayesian network approach, where BNs are built and validated employing data stored in DBs, and through a refinement process performed by the user. With this regard, DM is meant as a learning process more than a way to discover implicit correlations among data, as instead it is regarded in this work.…”

Section: Related Workmentioning

confidence: 99%

Homeland situation awareness through mining and fusing heterogeneous information from intelligence databases and field sensors

Digioia

Panzieri

2013

IJSSE

View full text Add to dashboard Cite

One of the most felt issues in the defence domain is that of having huge quantities of data stored in databases and acquired from field sensors without being able to infer information from them. Usually, databases are continuously updated with observations, and are related to heterogeneous data. Deep and continuous analysis on the data could mine useful correlations, explain relations existing among data and cue searches for further evidences. The solution to the problem addressed before seems to deal both with the domain of data mining and with the domain of high level data fusion. The focus of this paper is the definition of an architecture for a system adopting data mining techniques to adaptively discover clusters of information and relation among them, to classify observations acquired and to use the model of knowledge and the classification derived in order to assess situations, threats and refine the search for evidences.\ud Keywords: situation awareness; SAW; data mining; hidden Markov models; HMMs; agile modelling; intelligence databases; field sensors; data fusion; data clusters; information integration; classification; threats; defence industry; homeland securit

show abstract

Section: Related Workmentioning

confidence: 99%

Homeland situation awareness through mining and fusing heterogeneous information from intelligence databases and field sensors

Digioia

Panzieri

2013

IJSSE

View full text Add to dashboard Cite

show abstract

“…Bayesian networks (Pearl, 1988) are graphical representations of beliefs linked together with conditional probability distributions. This representation allows the potential solutions to be computationally evaluated against conditions and their relationships using hypothetical beliefs or real data (Stark, Farry, & Pfautz, 2012). The primary benefit of Bayesian networks in this context is that, unlike constrained argumentation networks, Bayesian networks do not dictate a particular structure.…”

Section: Supporting Quantitative Modelingmentioning

confidence: 99%

Incrementally Formalizing Graphical Models for Collaborative Operations Research

Stärk¹,

Roth

Farry

2013

Proceedings of the Human Factors and Ergonomics Society Annual

Self Cite

View full text Add to dashboard Cite

To address issues in enterprise systems, operations research (OR) analysts need to be able to understand, codify, and communicate various aspects of the issues, such as the environment's conditions and their relationships. Models are a natural way to capture that information, but they must be understandable to a variety of stakeholders involved in improving the enterprise. At the same time, the large amount of available information, such as event history and weather data, can easily overload analysts. To help analysts cope with data overload, it is useful for models to be accessible to computational tools that can provide data processing and visualization capabilities. To support both of those goals simultaneously, we describe an approach that supports the elicitation of qualitative insight from operations researchers and other relevant stakeholders and also provides avenues for computer software to perform semantic labeling and quantitative data processing. This approach directly supports an iterative OR process that satisfies the needs of multiple stakeholder communities, enabling initial qualitative relationships and hypotheses to be further investigated and justified with data-driven conclusions. Building on our previous experiences in knowledge acquisition and quantitative analysis, this paper outlines a new integrated workflow and a collection of graphical representation concepts for operations research and similar domains.

show abstract

“…Second, modeling failures incorporates biases of system designers, which are often never explicitly addressed in practice. Although recent research has looked into minimizing bias in model construction (Stark, Farry, & Pfautz, 2012), such approaches must be embedded in a system that is explicitly built to improve resilience. Also, designers of such tools do not always know how they will be used, especially in complex environments such as submarines.…”

Section: Introductionmentioning

confidence: 99%

Visualizations and Interaction Methods for Resilient Submarine Decision Support

Stärk

Woods

Farry

et al. 2012

Proceedings of the Human Factors and Ergonomics Society Annual

Self Cite

View full text Add to dashboard Cite

Submarine commanders must make decisions rapidly to carry out increasingly complex missions. However, the rate of information delivery has outpaced the capacity of the command and control systems that prioritize and filter it. Technology could help commanders filter through data to make decisions, but this decision support must be carefully engineered to support the development of resilient courses of action (COAs). This paper details our experiences applying resilience engineering to submarine decision support. It focuses on designing two features that are essential for a resilient decision support system: (1) user interaction with a decision support system, which blends the user's operational insights with the technical aspects of the decision support system; and (2) visual representations of trade-offs. The paper ends with a discussion of the lessons learned from this work and a set of recommendations for designing decision support systems.

show abstract

Mixed-initiative data mining with Bayesian networks

Cited by 3 publications

References 12 publications

Homeland situation awareness through mining and fusing heterogeneous information from intelligence databases and field sensors

Homeland situation awareness through mining and fusing heterogeneous information from intelligence databases and field sensors

Incrementally Formalizing Graphical Models for Collaborative Operations Research

Visualizations and Interaction Methods for Resilient Submarine Decision Support

Contact Info

Product

Resources

About