Fusion, propagation, and structuring in belief networks
Artificial Intelligence
Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Search-based methods to bound diagnostic probabilities in very large belief nets
Proceedings of the seventh conference (1991) on Uncertainty in artificial intelligence
Approximating probabilistic inference in Bayesian belief networks is NP-hard
Artificial Intelligence
Computers and Biomedical Research
Artificial intelligence: a modern approach
Artificial intelligence: a modern approach
An optimal approximation algorithm for Bayesian inference
Artificial Intelligence
The use of description logics in KBSE systems: experience report
ICSE '94 Proceedings of the 16th international conference on Software engineering
Weighing and Integrating Evidence for Stochastic Simulation in Bayesian Networks
UAI '89 Proceedings of the Fifth Annual Conference on Uncertainty in Artificial Intelligence
Efficient reasoning in qualitative probabilistic networks
AAAI'93 Proceedings of the eleventh national conference on Artificial intelligence
Inferring informational goals from free-text queries: a Bayesian approach
UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence
| Hi-index | 0.00 |
This paper introduces a computational framework for reasoning in Qualitative Belief Network (QBN) that derives its basis from inductive inference and reasoning. QBNs are essentially based on Bayesian Belief Networks (BBN), except that here the numerical probabilities of BBN are replaced by qualitative symbols. The relationships among the symbols provide a leeway to get good solutions with a qualitative approach to data utilization. The reasoning algebra is based on the usage of sign tables to propagate a belief through the QBN, in a guided approach to discover the causes in the causal relationships in QBN. This algorithm is also ideally suited to a distributed environment as it can absorb queries from multiple sources. The basis of this paper is the work done by Marek J. Druzdzel and the propagation algorithm that was proposed by him and Max Henrion for QBNs [5]. Their algorithm had problems in dealing with situations that might arise in normal circumstances e.g. the degrees of Belief in a particular event. In this paper, we have addressed the above issues and extended the reasoning algorithm by adding more levels in Belief by utilizing certain logical implications derived from basic rules of reasoning. Our algorithm also handles the issue of interactive processing and reasoning thereby making it capable of being used in a distributed platform. Given any data model, this approach helps in efficient reasoning of a solution which may not be directly evident from the singular belief in QBN. We have also implemented this algorithm to handle real life situations and the results thus obtained are in keeping with our expectations.