Saliency-Based Multilabel Linear Discriminant Analysis

Xu, Lei; Raitoharju, Jenni; Iosifidis, Alexandros; Gabbouj, Moncef

doi:10.1109/tcyb.2021.3069338

Cited by 23 publications

(12 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to evaluate the discriminating power of gut microbiomes’ multifractal dimensions in ages of infants, we used multifractal dimensions D ( q ) ( q from to 15 with step 0.2) of infants at 12 months and 4 months, when they were a baby, and the mothers’ gut microbiomes to discriminate by linear discriminant analysis (LDA) in R ( Xu et al., 2021 ). It consumed only 87 s to differentiate.…”

Section: Materials Methods and Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Microbial characterization based on multifractal analysis of metagenomes

Xie

Huang

Han

et al. 2023

Front. Cell. Infect. Microbiol.

View full text Add to dashboard Cite

IntroductionThe species diversity of microbiomes is a cutting-edge concept in metagenomic research. In this study, we propose a multifractal analysis for metagenomic research.Method and ResultsFirstly, we visualized the chaotic game representation (CGR) of simulated metagenomes and real metagenomes. We find that metagenomes are visualized with self-similarity. Then we defined and calculated the multifractal dimension for the visualized plot of simulated and real metagenomes, respectively. By analyzing the Pearson correlation coefficients between the multifractal dimension and the traditional species diversity index, we obtain that the correlation coefficients between the multifractal dimension and the species richness index and Shannon diversity index reached the maximum value when q = 0, 1, and the correlation coefficient between the multifractal dimension and the Simpson diversity index reached the maximum value when q = 5. Finally, we apply our method to real metagenomes of the gut microbiota of 100 infants who are newborn and 4 and 12 months old. The results show that the multifractal dimensions of an infant's gut microbiomes can distinguish age differences.Conclusion and DiscussionThere is self-similarity among the CGRs of WGS of metagenomes, and the multifractal spectrum is an important characteristic for metagenomes. The traditional diversity indicators can be unified under the framework of multifractal analysis. These results coincided with similar results in macrobial ecology. The multifractal spectrum of infants’ gut microbiomes are related to the development of the infants.

show abstract

Section: Materials Methods and Resultsmentioning

confidence: 99%

“…gut microbiomes to discriminate by linear discriminant analysis (LDA) in R (Xu et al, 2021). It consumed only 87 s to differentiate.…”

Section: Application Of Multifractal Dimension In Metagenomes To Infa...mentioning

confidence: 99%

Microbial characterization based on multifractal analysis of metagenomes

Xie

Huang

Han

et al. 2023

Front. Cell. Infect. Microbiol.

View full text Add to dashboard Cite

show abstract

“…As shown in Figure 1 , we use linear discriminant analysis (LDA) to reduce the dimensionality of the gene expression matrix. In order to solve a multilabel classification problem efficiently and effectively, we need not only to consider the correlation of class labels and features of each data item but also to take into account the different cardinalities of the classes ( Xu et al, 2021 ). The basic idea of LDA is to project the high-dimensional samples into the optimal discriminant vector space in order to extract the categorical information and compress the spatial dimensionality ( Guo et al, 2020 ).…”

Section: Methodsmentioning

confidence: 99%

“…The basic idea of LDA is to project the high-dimensional samples into the optimal discriminant vector space in order to extract the categorical information and compress the spatial dimensionality ( Guo et al, 2020 ). At the same time, the projection ensures that the samples have the maximum inter-class distance and the minimum intra-class distance in the new subspace, i.e., the samples have the best separability in this space ( Xu et al, 2021 ). For the input single-cell data matrix (the number of genes is m, the number of cells is n, the number of classes is k and the dimension after dimensionality reduction is d), it is experimentally verified that the best performance is achieved when d equals k-1.…”

Section: Methodsmentioning

confidence: 99%

TransCluster: A Cell-Type Identification Method for single-cell RNA-Seq data using deep learning based on transformer

et al. 2022

View full text Add to dashboard Cite

Recent advances in single-cell RNA sequencing (scRNA-seq) have accelerated the development of techniques to classify thousands of cells through transcriptome profiling. As more and more scRNA-seq data become available, supervised cell type classification methods using externally well-annotated source data become more popular than unsupervised clustering algorithms. However, accurate cellular annotation of single cell transcription data remains a significant challenge. Here, we propose a hybrid network structure called TransCluster, which uses linear discriminant analysis and a modified Transformer to enhance feature learning. It is a cell-type identification tool for single-cell transcriptomic maps. It shows high accuracy and robustness in many cell data sets of different human tissues. It is superior to other known methods in external test data set. To our knowledge, TransCluster is the first attempt to use Transformer for annotating cell types of scRNA-seq, which greatly improves the accuracy of cell-type identification.

show abstract

“…Dimension reduction methods can be also used for MLC [6][11] [21]. Xu et al [45] proposed a probabilistic class saliency estimation approach for calculating the projection matrix in linear discriminant analysis for dimension reduction. Decision tree can be built recursively based on a multi-labelentropy based information gain criterion [14].…”

Section: Related Workmentioning

confidence: 99%

Robust fused hypergraph neural networks for multi-label classification

Wang

Yang

et al. 2023

IFS

View full text Add to dashboard Cite

Deep neural networks have been adopted in multi-label classification for their excellent performance, however, existing methods fail to comprehensively utilize the high-order correlations between instances and the high-order correlations between labels, and these methods are difficult to deal with label noise effectively. We propose a novel end-to-end deep framework named Robust Fused Hypergraph Neural Networks for Multi-Label Classification (RFHNN), which can effectively utilize the two kinds of high-order correlations and adopt them to mitigate the impact of label noise. In RFHNN, Hypergraph Neural Networks (HNNs) are adopted to mine and utilize the high-order correlations of the instances in the feature space and the label space respectively. The high-order correlations of the instances can not only improve the accuracy of the classification and the discrimination of the proposed model, but also lay the foundation for the subsequent noise correction module. Meanwhile, a hypergraph construction method based on the Apriori algorithm is proposed to realize Hypergraph Neural Networks (HNNs), which can mine robust second-order and high-order label correlations effectively. Effective classifiers are learned based on the correlations between the labels, which will not only improve the accuracy of the model, but can also enhance the subsequent noise correction module. In addition, we have designed a noise correction module in the networks. With the help of the high-order correlations among the instances and the effective classifier, the framework can effectively correct the noise and improve the robustness of the model. Extensive experimental results on datasets demonstrate that our proposed approach is better than the state-of-the-art multi-label classification algorithms. When dealing with the multi-label training datasets with noise in the label space, our proposed method also has great performance.

show abstract

Saliency-Based Multilabel Linear Discriminant Analysis

Cited by 23 publications

References 62 publications

Microbial characterization based on multifractal analysis of metagenomes

Microbial characterization based on multifractal analysis of metagenomes

TransCluster: A Cell-Type Identification Method for single-cell RNA-Seq data using deep learning based on transformer

Robust fused hypergraph neural networks for multi-label classification

Contact Info

Product

Resources

About