An interactive approach for Bayesian network learning using domain/expert knowledge

Masegosa, Andrés R.; Moral, Serafı́n

doi:10.1016/j.ijar.2013.03.009

Cited by 46 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meganck et al (2006) compare different utility functions for single-vertex interventions, but do not address algorithmic questions of efficiently calculating optima of the utility functions. In the Bayesian setting, Tong and Koller (2001) and Masegosa and Moral (2013) uses entropy-based utility functions. While the approach of Tong and Koller (2001) only interacts with the system under investigation, the approach of Masegosa and Moral (2013) use (error-free) expert knowledge.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Two optimal strategies for active learning of causal models from interventional data

Hauser

Bühlmann

2014

International Journal of Approximate Reasoning

126

View full text Add to dashboard Cite

From observational data alone, a causal DAG is only identifiable up to Markov equivalence. Interventional data generally improves identifiability; however, the gain of an intervention strongly depends on the intervention target, that is, the intervened variables. We present active learning (that is, optimal experimental design) strategies calculating optimal interventions for two different learning goals. The first one is a greedy approach using single-vertex interventions that maximizes the number of edges that can be oriented after each intervention. The second one yields in polynomial time a minimum set of targets of arbitrary size that guarantees full identifiability. This second approach proves a conjecture of Eberhardt (2008) indicating the number of unbounded intervention targets which is sufficient and in the worst case necessary for full identifiability. In a simulation study, we compare our two active learning approaches to random interventions and an existing approach, and analyze the influence of estimation errors on the overall performance of active learning.

show abstract

Section: Introductionmentioning

confidence: 99%

“…In the Bayesian setting, Tong and Koller (2001) and Masegosa and Moral (2013) uses entropy-based utility functions. While the approach of Tong and Koller (2001) only interacts with the system under investigation, the approach of Masegosa and Moral (2013) use (error-free) expert knowledge.…”

Section: Introductionmentioning

confidence: 99%

Two optimal strategies for active learning of causal models from interventional data

Hauser

Bühlmann

2014

International Journal of Approximate Reasoning

126

View full text Add to dashboard Cite

show abstract

“…The qualitative part (i.e., the structure of the Bayesian network) consists of a directed acyclic graph (DAG) where the nodes correspond to the variables in the domain problem and the edges between two variables correspond to direct probabilistic dependencies. On the other hand, the quantitative part consists of the specification of the conditional probability distributions that are stored in the nodes of the network [28]. DAG describes the relationship between attributes and consists of nodes and arcs, where each arc describes a probabilistic dependence.…”

Section: B Bayesian Networkmentioning

confidence: 99%

Improved-Node-Probability Method for Decision Making in Priority Determination of Village Development Proposed Program

Trisnawarman¹,

Hartati²,

Winarko³

et al. 2017

ijacsa

View full text Add to dashboard Cite

Abstract-This research proposes a new method, the probability of nodes (NP) and the cumulative frequency of indicators within the framework of Bayesian networks to calculate the weight of participation. This method uses the PLS-PM approach to examine the relationship structure of participatory factors and estimate latent variables. Data were collected using questionnaires involving participants offering proposals, the village residents themselves. The participation factors identified in this research were divided into two categories, namely, internal factors (abilities) and external factors (motivation). The internal factors included gender, age, education, occupation, and income, while the external factors included motivation relating to economic, political, sociocultural, norm-related, and knowledge-related issues. Moreover, there are three factors directly affecting the level of participation, they are: the level of attendance in meetings, participation in giving suggestions, and involvement in decision making. The test results showed that the application of participation weight in decision making priority of proposal of village development program give change of final rank of decision with test result as: recall 50%, precision 80% and accuracy 50%.

show abstract

“…Cano et al [8] and Flores et al [35] proposed an interactive BN structure learning approach that iteratively queries the domain expert about the reliability of learnt edges. This interactive paradigm can be easily applied to help elicit edges' monotonic influences from experts and suggests that the theoretical method presented in this paper can be applied in practice (we do not use the interactive paradigm here because in our experiments we assume that any edges with monotonic influences are known and we simulate errors made by experts).…”

Section: Related Workmentioning

confidence: 99%

An empirical study of Bayesian network parameter learning with monotonic influence constraints

Zhou

Fenton

Zhu

2016

Decision Support Systems

View full text Add to dashboard Cite

A B S T R A C TLearning the conditional probability table (CPT) parameters of Bayesian networks (BNs) is a key challenge in real-world decision support applications, especially when there are limited data available. A conventional way to address this challenge is to introduce domain knowledge/expert judgments that are encoded as qualitative parameter constraints. In this paper we focus on a class of constraints which is naturally encoded in the edges of BNs with monotonic influences. Experimental results indicate that such monotonic influence constraints are widespread in practical BNs (all BNs used in the study contain such monotonic influences). To exploit expert knowledge about such constraints we have developed an improved constrained optimization algorithm, which achieves good parameter learning performance using these constraints, especially when data are limited. Specifically, this algorithm outperforms the previous state-of-the-art and is also robust to errors in labelling the monotonic influences. The method is applied to a real world medical decision support BN where we had access to expert-provided constraints and real hospital data. The results suggest that incorporating expert judgments about monotonic influence constraints can lead to more accurate BNs for decision support and risk analysis.

show abstract

An interactive approach for Bayesian network learning using domain/expert knowledge

Cited by 46 publications

References 20 publications

Two optimal strategies for active learning of causal models from interventional data

Two optimal strategies for active learning of causal models from interventional data

Improved-Node-Probability Method for Decision Making in Priority Determination of Village Development Proposed Program

An empirical study of Bayesian network parameter learning with monotonic influence constraints

Contact Info

Product

Resources

About