Learning Bayesian networks using the constrained maximum a posteriori probability method

Yu, Yang; Gao, Xiaoguang; Guo, Zhiwei; Chen, Daqing

doi:10.1016/j.patcog.2019.02.006

Cited by 28 publications

(18 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different methods can be used to quantify the similarity between the two signals including correlation, maximum a posteriori probability estimation, and dynamic time wrapping. In this study, a correlation function was utilized due to its simplicity of integration [24,25]. So, the symbol with the highest correlation with the received signal was identified.…”

Section: Methodsmentioning

confidence: 99%

An alternative method of biomedical signal transmission through the GSM voice channel

Dehghanojamahalleh¹,

Kaya²

2020

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

In this work, a new solution for online and accurate biomedical data transmission is presented. For this purpose, a global system for mobile (GSM) communication voice channel is, for the first time, used as a communication link between the patient and healthcare provider. Biomedical signals are converted into speech-like signals before being transferred over a GSM voice channel. On the receiver side, speech-like symbols are stored in a symbols bank, and constructed using random stochastic signals. On the receiver end, the index of the symbol with the most similarity to the received signal is selected as the identified sample. This method enables us to communicate with an accuracy of 99.8 % at a transfer rate of 110 samples per second and signal-to-noise ratio (SNR) of 10. By utilizing a GSM voice channel, any voice channel, such as a cell phone, can be used for data transmission. The transmitted signal is encoded; therefore, the connection is secured. GSM technology has benefits such as availability, reliability, and robustness. Additionally, GSM can be used as a backup or service for transmitting vital physiological signals in emergency situations (e.g. in an ambulance). This technology can also be used to transmit other physiological signals as well as nonphysiological generic data.

show abstract

Section: Methodsmentioning

confidence: 99%

An alternative method of biomedical signal transmission through the GSM voice channel

Dehghanojamahalleh¹,

Kaya²

2020

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

show abstract

“…-Constrained MAP approach has also been proposed by Yang et al (2019) is available yet; practical recommendations on how to implement them and their limitations is currently lacking in the literature.…”

Section: Combining Scarce Data and Expert Judgementsmentioning

confidence: 99%

Uncertainties in conditional probability tables of discrete Bayesian Belief Networks: A comprehensive review

Rohmer

2020

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

Discrete Bayesian Belief Network (BBN) has become a popular method for the analysis of complex systems in various domains of application. One of its pillar is the specification of the parameters of the probabilistic dependence model (i.e. the cause-effect relation) represented via a Conditional Probability Table (CPT). Depending on the available data (observations, prior knowledge, expert-based information, etc.), CPTs can be populated in different manners, i.e. different assumptions can be made and different methods are available, which might lead to uncertain BBN-based results. Through an extensive review study of the past ten years, we aim at addressing three questions related to the CPT uncertainties. First, we show how to constrain these uncertainties either using elicitation of expert inputs, or using a combination of scarce data and expert-derived information. Second, we show how to integrate these uncertainties in the BBN-based analysis through propagation procedures either using probabilities or imprecise probabilities within the setting of credal or evidential networks. Finally, we show how to test the robustness of the BBN-based results to these uncertainties via sensitivity analysis specifically dedicated to BBNs. A special care was paid to describe the best practices for the implementation of the reviewed methods and the remaining gaps.

show abstract

“…The combination is the focus of this article, with the peculiarity that the human expert is replaced by the expert machine, i.e. the CBR [7,9,14,15].…”

Section: Knowledge Engineering Bayesian Network (Kebn)mentioning

confidence: 99%

Integrating CBR with Data in Bayesian Networks for Decision Making in an Echelon Supply Chain Distribution Solution

Castro¹,

Ochoa-Zezzatti²,

Sánchez³

et al. 2019

RCS

View full text Add to dashboard Cite

When developing a causal probabilistic model, that is, a Bayesian network (BN), it is common to incorporate expert knowledge of factors that are important for decision analysis, but there are models where historical data is not available or difficult to obtain, or it is difficult to have a human expert nearby to help. This document explains how data is developed from a discrete/continuous simulated variable through a BN and mixed integerlinear programming (MILP), and the impact of this variable is measured as an important element for the decision-making model. Consider as an additional expert variable. The CBR model and the variable in question is contextualized to support in the decision-making process in a supply chain through two stages, the first is considered multiple factories, with multiple distribution centers (DC) and second, from the multiple distribution centers as it reaches multiple points of sale. As a design of a decision support system for the construction of a supply chain network (SCN) for a range of multiple end products, as well as the determination of factories and distribution centers, it also helps in the design of the distribution network strategy that satisfies all the capacities and requirements of demand of the product imposed through the points of sale. At the end of the work, an evaluation of the performance of two Bayesian networks is carried out, where one of them represents the incorporation of the expert variable using two methods, one of them the receiver operating characteristic (ROC) curve and two a method proposed by Constantinou et al. [2]., Where in both cases the Bayesian network gave a better performance with the expert variable.

show abstract

Learning Bayesian networks using the constrained maximum a posteriori probability method

Cited by 28 publications

References 26 publications

An alternative method of biomedical signal transmission through the GSM voice channel

An alternative method of biomedical signal transmission through the GSM voice channel

Uncertainties in conditional probability tables of discrete Bayesian Belief Networks: A comprehensive review

Integrating CBR with Data in Bayesian Networks for Decision Making in an Echelon Supply Chain Distribution Solution

Contact Info

Product

Resources

About