Efficient Heuristics for Structure Learning of k-Dependence Bayesian Classifier

Liu, Yang; Wang, Limin; Sun, Minghui

doi:10.3390/e20120897

Cited by 7 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further demonstrate the performance of UKDB over KDB, we employ the goal difference (GD) [ 19 , 21 ]. Suppose there are two classifiers A and B , the value of

can be computed as follow:

where

is the datasets,

and

represent the number of datasets on which A performs better or worse than B , respectively.…”

Section: Resultsmentioning

confidence: 99%

“…Since KDB can be extended to describe dependence relationships of arbitrary degree and thus demonstrates its flexibility, researchers proposed many important refinements to improve its performance [ 18 , 19 , 20 , 21 ]. Pernkopf and Bilmes [ 22 ] proposed a greedy heuristic strategy to determine the attribute order by comparing

where

ranks higher than

in the order, i.e.,

.…”

Section: Bayesian Network and Markov Blanketmentioning

confidence: 99%

See 1 more Smart Citation

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

“…To further demonstrate the performance of UKDB over KDB, we employ the goal difference (GD) [ 19 , 21 ]. Suppose there are two classifiers A and B , the value of

can be computed as follow:

where

is the datasets,

and

represent the number of datasets on which A performs better or worse than B , respectively.…”

Section: Resultsmentioning

confidence: 99%

where

ranks higher than

in the order, i.e.,

.…”

Section: Bayesian Network and Markov Blanketmentioning

confidence: 99%

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

“…To further compare the performance of UKDB with other mentioned algorithms in terms of data size, the Goal Difference (GD) [ 51 , 52 ] was introduced. Suppose for two classifiers A , B , we compute the value of GD as follows:

where

represents the datasets for comparison and

and

are respectively the numbers of datasets on which the classification performance of A is better or worse than that of B .…”

Section: Resultsmentioning

confidence: 99%

Universal Target Learning: An Efficient and Effective Technique for Semi-Naive Bayesian Learning

Gao

Hua²,

Wang

et al. 2019

Entropy

Self Cite

View full text Add to dashboard Cite

To mitigate the negative effect of classification bias caused by overfitting, semi-naive Bayesian techniques seek to mine the implicit dependency relationships in unlabeled testing instances. By redefining some criteria from information theory, Target Learning (TL) proposes to build for each unlabeled testing instance P the Bayesian Network Classifier BNC P , which is independent and complementary to BNC T learned from training data T . In this paper, we extend TL to Universal Target Learning (UTL) to identify redundant correlations between attribute values and maximize the bits encoded in the Bayesian network in terms of log likelihood. We take the k-dependence Bayesian classifier as an example to investigate the effect of UTL on BNC P and BNC T . Our extensive experimental results on 40 UCI datasets show that UTL can help BNC improve the generalization performance.

show abstract

“…Structure of classifier in [11] is the source of our network. Our network design is shown in detail in Fig 4 . Our discriminator may be thought of as a way to reduce the pattern classification loss between intensities picture and events.…”

Section: Classifier Structurementioning

confidence: 99%

Range Imaging and Video Generation using Generative Adversarial Network

2021

JBSHA

View full text Add to dashboard Cite

Latency, high temporal pixel density, and dynamic range are just a few of the benefits of event camera systems over conventional camera systems. Methods and algorithms cannot be applied directly because the output data of event camera systems are segments of synchronization events and experiences rather than precise pixel intensities. As a result, generating intensity photographs from occurrences for other functions is difficult. We use occurrence camera-based contingent deep convolutional connections to establish images and videos from a variable component of the occasion stream of data in this journal article. The system is designed to replicate visuals based on spatio-temporal intensity variations using bundles of spatial coordinates of occurrences as input data. The ability of event camera systems to produce High Dynamic Range (HDR) pictures even in exceptional lighting circumstances, as well as non-blurry pictures in rapid motion, is demonstrated. Furthermore, because event cameras have a transient response of about 1 s, the ability to generate very increased frame rate video content has been evidenced, conceivably up to 1 million arrays per second. The implementation of the proposed algorithms are compared to density images recorded onto a similar gridline in the image of events based on the application of accessible primary data obtained and synthesized datasets generated by the occurrence camera simulation model.

show abstract

Efficient Heuristics for Structure Learning of k-Dependence Bayesian Classifier

Cited by 7 publications

References 19 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

Universal Target Learning: An Efficient and Effective Technique for Semi-Naive Bayesian Learning

Range Imaging and Video Generation using Generative Adversarial Network

Contact Info

Product

Resources

About