Target Learning: A Novel Framework to Mine Significant Dependencies for Unlabeled Data

Wang, Limin; Chen, Shenglei; Mammadov, Musa

doi:10.1007/978-3-319-93034-3_9

Cited by 10 publications

(13 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CMI can weigh the conditional dependency between Sex and Pregnant , but cannot discriminately weigh the dependencies when these two attributes take different values. Target learning takes each testing instance

as a target and tries to mine the dependence relationships between these attribute values [ 16 ]. From Equations ( 1 ) and ( 5 ), we have the following equations:

where

…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

“…The definitions of MI and CMI are measures of the average dependence between attributes implicated in the training data. In contrast to those, local mutual information (LMI)

and conditional local mutual information (CLMI)

can weigh the direct dependence and conditional dependence relationships between attribute values implicated in each instance [ 16 , 24 ]. Similarly, we sort the attribute values by comparing the sum of CLMI (SCLMI) and

.…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

“…UKDB

learned from training data

describes the general conditional dependencies, thus UKDB

corresponds to the domain knowledge that may be suitable for most cases. In contrast, UKDB

learned from testing instance

describes local conditional dependencies with uncertainty because all class labels are considered, thus UKDB

corresponds to the personalized knowledge that may be suitable for

only [ 16 ].…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

“…To meet the needs of experts in machine learning or in medicine, common people (including patients and their relatives) and the problem of breast cancer mentioned above, we propose a novel sorting strategy and extend KDB from a single restricted network to unrestricted ensemble networks, i.e., unrestricted k -dependence Bayesian classifier (UKDB), in terms of Markov blanket analysis and target learning. Target learning [ 16 ] is a framework that takes each unlabeled testing instance

as a target and builds a specific Bayesian model BNC

to complement BNC

learned from training data

.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Discriminative Structure Learning of Bayesian Network Classifiers from Training Dataset and Testing Instance

Wang

Liu

Mammadov

et al. 2019

Entropy

Self Cite

View full text Add to dashboard Cite

Over recent decades, the rapid growth in data makes ever more urgent the quest for highly scalable Bayesian networks that have better classification performance and expressivity (that is, capacity to respectively describe dependence relationships between attributes in different situations). To reduce the search space of possible attribute orders, k-dependence Bayesian classifier (KDB) simply applies mutual information to sort attributes. This sorting strategy is very efficient but it neglects the conditional dependencies between attributes and is sub-optimal. In this paper, we propose a novel sorting strategy and extend KDB from a single restricted network to unrestricted ensemble networks, i.e., unrestricted Bayesian classifier (UKDB), in terms of Markov blanket analysis and target learning. Target learning is a framework that takes each unlabeled testing instance P as a target and builds a specific Bayesian model Bayesian network classifiers (BNC) P to complement BNC T learned from training data T . UKDB respectively introduced UKDB P and UKDB T to flexibly describe the change in dependence relationships for different testing instances and the robust dependence relationships implicated in training data. They both use UKDB as the base classifier by applying the same learning strategy while modeling different parts of the data space, thus they are complementary in nature. The extensive experimental results on the Wisconsin breast cancer database for case study and other 10 datasets by involving classifiers with different structure complexities, such as Naive Bayes (0-dependence), Tree augmented Naive Bayes (1-dependence) and KDB (arbitrary k-dependence), prove the effectiveness and robustness of the proposed approach.

show abstract

as a target and tries to mine the dependence relationships between these attribute values [ 16 ]. From Equations ( 1 ) and ( 5 ), we have the following equations:

where

…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

“…The definitions of MI and CMI are measures of the average dependence between attributes implicated in the training data. In contrast to those, local mutual information (LMI)

and conditional local mutual information (CLMI)

.…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

“…UKDB

learned from training data

describes the general conditional dependencies, thus UKDB

corresponds to the domain knowledge that may be suitable for most cases. In contrast, UKDB

learned from testing instance

describes local conditional dependencies with uncertainty because all class labels are considered, thus UKDB

corresponds to the personalized knowledge that may be suitable for

only [ 16 ].…”

Section: The Ukdb Algorithmmentioning

confidence: 99%

as a target and builds a specific Bayesian model BNC

to complement BNC

learned from training data

.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Discriminative Structure Learning of Bayesian Network Classifiers from Training Dataset and Testing Instance

Wang

Liu

Mammadov

et al. 2019

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is even more difficult for BSE or FSS to select the conditional dependencies. For example, the network topology of KDB consists of nk − k 2 2 − k 2 conditional dependencies [21]. If BSE or FSS evaluate them one by one to identify those relatively non-significant ones, the high computational overheads is almost unbearable and few approaches are proposed to address this issue.…”

Section: Introductionmentioning

confidence: 99%

Structure Learning of Bayesian Network Based on Adaptive Thresholding

Zhang

Wang

Duan

et al. 2019

Entropy

Self Cite

View full text Add to dashboard Cite

Direct dependencies and conditional dependencies in restricted Bayesian network classifiers (BNCs) are two basic kinds of dependencies. Traditional approaches, such as filter and wrapper, have proved to be beneficial to identify non-significant dependencies one by one, whereas the high computational overheads make them inefficient especially for those BNCs with high structural complexity. Study of the distributions of information-theoretic measures provides a feasible approach to identifying non-significant dependencies in batch that may help increase the structure reliability and avoid overfitting. In this paper, we investigate two extensions to the k-dependence Bayesian classifier, MI-based feature selection, and CMI-based dependence selection. These two techniques apply a novel adaptive thresholding method to filter out redundancy and can work jointly. Experimental results on 30 datasets from the UCI machine learning repository demonstrate that adaptive thresholds can help distinguish between dependencies and independencies and the proposed algorithm achieves competitive classification performance compared to several state-of-the-art BNCs in terms of 0–1 loss, root mean squared error, bias, and variance.

show abstract

Stochastic optimization for bayesian network classifiers

Ren

Wang

et al. 2022

Appl Intell

View full text Add to dashboard Cite

Target Learning: A Novel Framework to Mine Significant Dependencies for Unlabeled Data

Cited by 10 publications

References 16 publications

Discriminative Structure Learning of Bayesian Network Classifiers from Training Dataset and Testing Instance

Discriminative Structure Learning of Bayesian Network Classifiers from Training Dataset and Testing Instance

Structure Learning of Bayesian Network Based on Adaptive Thresholding

Stochastic optimization for bayesian network classifiers

Contact Info

Product

Resources

About