Multi-Label Dimensionality Reduction

Ye, Jieping; Sun, Liang

doi:10.1201/b16017

Cited by 27 publications

(40 citation statements)

References 165 publications

(412 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Similar to many machine learning and data mining tasks, multilabel learning also suffers from the curse of dimensionality [7]. In many multilabel learning applications such as text categorization, the dimensionality of instance space and label space can be very high.…”

Section: Related Workmentioning

confidence: 99%

“…first-order strategy, second-order strategy, and high-order strategy based on the order of label correlations that the learning algorithms have considered. Recently, several dimensionality reduction algorithms have been proposed for multilabel learning [7], [20]. Many of these algorithms attempt to solve the curse of dimensionality problem meanwhile capture correlations among labels.…”

Section: Related Workmentioning

confidence: 99%

“…x W W (7) In this paper, we utilize a log function to describe the energy function of MLHN. Therefore, the energy function is defined in (8), where |E| is the total number of hyperedges in MLHN, wji is the weight of the label i of hyperedge ej, I(x,yi; ej) is a matching function which is defined in (9).…”

Section: Multilabel Hypernetworkmentioning

confidence: 99%

“…The generality of multilabel classification makes it challenging to solve. The main challenge of multilabel classification lies in how to effectively and efficiently exploit correlations among labels [7]. A major difference between multilabel classification and traditional binary or multi-class classification is that labels in multilabel classification are not mutually exclusive but may be correlated.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Multilabel Classification via Co-Evolutionary Multilabel Hypernetwork

Sun¹,

Lee²,

Jin³

2016

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

Multilabel classification is prevalent in many real-world applications where data instances may be associated with multiple labels simultaneously. In multilabel classification, exploiting label correlations is an essential but nontrivial task. Most of the existing multilabel learning algorithms are either ineffective or computational demanding and less scalable in exploiting label correlations. In this paper, we propose a co-evolutionary multilabel hypernetwork (Co-MLHN) as an attempt to exploit label correlations in an effective and efficient way. To this end, we firstly convert the traditional hypernetwork into a multilabel hypernetwork (MLHN) where label correlations are explicitly represented. We then propose a co-evolutionary learning algorithm to learn an integrated classification model for all labels. The proposed Co-MLHN exploits arbitrary order label correlations and has linear computational complexity with respect to the number of labels. Empirical studies on a broad range of multilabel data sets demonstrate that Co-MLHN achieves competitive results against state-of-the-art multilabel learning algorithms, in terms of both classification performance and scalability with respect to the number of labels.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Multilabel Hypernetworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multilabel Classification via Co-Evolutionary Multilabel Hypernetwork

Sun¹,

Lee²,

Jin³

2016

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

show abstract

“…Li et al [33] present a novel multi-label dimensionality reduction using the variable pairwise constraints. A more comprehensive review of multi-label dimensionality reduction as well as multi-label learning algorithms can be found in [34,35].…”

Section: Introductionmentioning

confidence: 99%

Learning shared subspace for multi-label dimensionality reduction via dependence maximization

Shu

Lai

et al. 2015

Neurocomputing

View full text Add to dashboard Cite

Latent vector autoregressive modeling and feature analysis of high dimensional and noisy data from dynamic systems

Qin

2022

AIChE Journal

View full text Add to dashboard Cite

In this article, a novel latent vector autoregressive (LaVAR) modeling algorithm with a canonical correlation analysis (CCA) objective is proposed to estimate a fully‐interacting reduced‐dimensional dynamic model. This algorithm is an advancement of the dynamic inner canonical correlation analysis (DiCCA) algorithm, which builds univariate latent autoregressive models that are noninteracting. The dynamic latent variable scores of the proposed algorithm are guaranteed to be orthogonal with a descending order of predictability, retaining the properties of DiCCA. Further, the LaVAR‐CCA algorithm solves multiple latent variables simultaneously with a statistical interpretation of the profile likelihood. The Lorenz oscillator with noisy measurements and an application case study on an industrial dataset are used to illustrate the superiority of the proposed algorithm. The reduced‐dimensional latent dynamic model has numerous potential applications for prediction, feature analysis, and diagnosis of systems with rich measurements.

show abstract

Multi-Label Dimensionality Reduction

Cited by 27 publications

References 165 publications

Multilabel Classification via Co-Evolutionary Multilabel Hypernetwork

Multilabel Classification via Co-Evolutionary Multilabel Hypernetwork

Learning shared subspace for multi-label dimensionality reduction via dependence maximization

Latent vector autoregressive modeling and feature analysis of high dimensional and noisy data from dynamic systems

Contact Info

Product

Resources

About