A framework for building knowledge-bases under uncertainty

Santos, Eugene

doi:10.1080/095281399146571

Cited by 54 publications

(35 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…BKBs have been extensively studied both theoretically [33,34] and for use in knowledge engineering [25] in a wide variety of domains such as space shuttle engine diagnosis [3,25], medical information processing [20], and freshwater aquarium maintenance [33]. BKBs provide a highly flexible and intuitive representation following a basic "if-then" structure in conjunction with probability theory.…”

Section: Background: Terms and Conceptsmentioning

confidence: 99%

“…Various approaches to reasoning with Bayesian Networks include A* search, stochastic simulation, integer programming, and message passing [20,26,28,21,27,33] …”

Section: Uncertaintymentioning

confidence: 99%

“…We also recommend the migration to Bayesian Knowledge-Bases as the knowledge representation [33,34]. BKBs do not require a complete instantiation of all conditional probabilities since BKBs are rule-based in nature.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Adversarial Intent Inference for Predictive Battlespace Awareness

Santos¹,

Eugene²

2005

View full text Add to dashboard Cite

REPORT DOCUMENTATION PAGEForm Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and ADVERSARIAL INTENT INFERENCE FOR PREDICTIVE BATTLESPACE AWARENESS AUTHOR(S)Eugene Santos, Jr. FUNDING NUMBERSG -F30602-01-1-0595 PE -62702F PR -558B TA -II WU -09 PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)University of Connecticut Office for Sponsored Programs 438 Whitney Road, Extn, Unit 1133 Storrs CT 06269-2938 PERFORMING ORGANIZATION REPORT NUMBERN/A SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES)AFRL 12b. DISTRIBUTION CODE ABSTRACT (Maximum 200 Words)Designed and developed a cognitive architecture to model the adversary which forms the basis for the Adversary Intent Inferencing (AII) Module. Designed, developed, and implemented the AII Module based on both Bayesian Networks and Bayesian Knowledge Bases for adversarial modeling, course of action prediction, explanation, and inference of adversary intent. All functioning in both wintel and Unix environments. Integrated AII module into prototype system for modeling and predicting the adversary based on the Battle of Khafji scenario. Integrated AII module in Force Structure Simulation wargaming system and demonstrated emergent behavior when Red force behavior is varied (AFRL/IFTC). Developed Dynamico tool for constructing Bayesian fragments and templates into libraries for use by the AII -resulted in a MS Thesis. AII transitioned into Phase I and Phase II SBIR Projects under Securboration -"Emergent Adversarial Modeling System". NUMBER OF PAGES 14. SUBJECT TERMS Final Project SummaryGoal: Design and develop computational framework for adversarial modeling and intent inferencing for decision support Approach: Dynamically capture and predict enemy interests, goals, rationale, and courses of action under uncertainty through machine learning and Bayesian networks To achieve adversarial intent inferencing requires the ability to (1) fuse information (observables) from sensors and intelligence sources regarding the adversary, (2) infer adversary intent and goals, and (3) predict adversary courses of action (COA). In total, adversary intent inferencing (AII) provides these three key functions while also taking into consideration a number of utility issues:• AII must be able to explain the basis of its predictions; why is the adversary pursuing a predicted goal? What is driving the adversary to pursue these COAs? Must be able to model and take into account many factors including soft factors such as political environment, personality ...

show abstract

Section: Background: Terms and Conceptsmentioning

confidence: 99%

“…Various approaches to reasoning with Bayesian Networks include A* search, stochastic simulation, integer programming, and message passing [20,26,28,21,27,33] …”

Section: Uncertaintymentioning

confidence: 99%

See 1 more Smart Citation

Adversarial Intent Inference for Predictive Battlespace Awareness

Santos¹,

Eugene²

2005

View full text Add to dashboard Cite

show abstract

“…A Bayesian knowledge base (BKB) is a probabilistic knowledge representation meeting the preceding qualities [4]. A BKB supports theoretically sound and consistent probabilistic inference Ñ even with incomplete knowledge Ñ with the intuitiveness of Òif-thenÓ rule specification.…”

Section: Bayesian Knowledge Basesmentioning

confidence: 99%

Active User Interfaces For Building Decision-Theoretic Systems

Brown

Santos²,

Banks³

1999

Self Cite

View full text Add to dashboard Cite

Knowledge elicitation/acquisition continues to be a bottleneck to constructing decisiontheoretic systems. Methodologies and techniques for incremental elicitation/acquisition of knowledge especially under uncertainty in support of users' current goals is desirable. This paper presents PESKI, a probabilistic expert system development environment. PESKI provides users with a highly interactive and integrated suite of intelligent knowledge engineering tools for decision-theoretic systems. From knowledge acquisition, data mining, and verification and validation to a distributed inference engine for querying knowledge, PESKI is based on the concept of active user interfaces --actuators to the human-machine interface. PESKI uses a number of techniques to reduce the inherent complexity of developing a cohesive, real-world knowledge-based system. This is accomplished by providing multiple communication modes for human-computer interaction and the use of a knowledge representation endowed with the ability to detect problems with the knowledge acquired and alert the user to these possible problems. We discuss PESKI's use of these intelligent assistants to help users with the acquisition of knowledge especially in the presence of uncertainty.

show abstract

“…In brief, we first learn subsystem dynamics through a probabilistic graphical model called Bayesian Knowledge Bases (BKBs) [8] from observations on each subsystem. Then we fuse these BKBs into one BKB via the BKB fusion algorithm [9], which includes interactions among subsystems both probabilistically and structurally sound.…”

Section: Introductionmentioning

confidence: 99%

Automatic Emergence Detection in Complex Systems

Santos

Zhao

2017

Complexity

Self Cite

View full text Add to dashboard Cite

Complex systems consist of multiple interacting subsystems, whose nonlinear interactions can result in unanticipated (emergent) system events. Extant systems analysis approaches fail to detect such emergent properties, since they analyze each subsystem separately and arrive at decisions typically through linear aggregations of individual analysis results. In this paper, we propose a quantitative definition of emergence for complex systems. We also propose a framework to detect emergent properties given observations of its subsystems. This framework, based on a probabilistic graphical model called Bayesian Knowledge Bases (BKBs), learns individual subsystem dynamics from data, probabilistically and structurally fuses said dynamics into a single complex system dynamics, and detects emergent properties. Fusion is the central element of our approach to account for situations when a common variable may have different probabilistic distributions in different subsystems. We evaluate our detection performance against a baseline approach (Bayesian Network ensemble) on synthetic testbeds from UCI datasets. To do so, we also introduce a method to simulate and a metric to measure discrepancies that occur with shared/common variables. Experiments demonstrate that our framework outperforms the baseline. In addition, we demonstrate that this framework has uniform polynomial time complexity across all three learning, fusion, and reasoning procedures.

show abstract

A framework for building knowledge-bases under uncertainty

Cited by 54 publications

References 19 publications

Adversarial Intent Inference for Predictive Battlespace Awareness

Adversarial Intent Inference for Predictive Battlespace Awareness

Active User Interfaces For Building Decision-Theoretic Systems

Automatic Emergence Detection in Complex Systems

Contact Info

Product

Resources

About