Learning matrix factorization with scalable distance metric and regularizer

Wang, Shiping; Zhang, Yunhe; Lin, Xincan; Su, Lichao; Xiao, Guobao; Zhu, William; Shi, Yiqing

doi:10.1016/j.neunet.2023.01.034

Cited by 7 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under hypoxic conditions, the LDH content in the cytosol increases, cells undergo morphological changes such as crumpling and deformation, and the cell structure, i.e., the integrity of the cell membrane and normal cell morphology, is damaged. Te results of the present experimental study showed that hypoxia caused reduced membrane integrity, destruction of cell morphology and structure, and apoptotic and necrotic morphology in H9c2 rat cardiomyocytes, which is consistent with the results of the previous studies [26,27]. Apoptosis is a programmed cell death that is complexly regulated by multiple genes and involves activation, expression, and regulation of intracellular-related factors, which have important roles in immune defense and cell damage in animal organisms [28,29].…”

Section: Discussionsupporting

confidence: 93%

The Effect of Lycium ruthenicum Murry Anthocyanins on the Apoptosis and Proliferation of H9c2 Cell Induced by Hypoxia

Li,

Wang,

et al. 2023

Journal of Food Biochemistry

View full text Add to dashboard Cite

Based on the strong antioxidant activity of anthocyanins, this study was aimed to explore the potential influence of Lycium ruthenicum Murry anthocyanins (LRMA) on hypoxia-induced apoptosis and proliferation of H9c2 rat cardiomyocytes. The cell viability was first tested after exposure to normoxia, hypoxia, and hypoxia + LRMA. After being incubated for 12 h under normoxia and 24 h under hypoxia in 50 μg/mL LRMA or 100 μM Salidroside, respectively, LDH content, cell morphology, and cell cycle progression were analyzed. In addition, cellular apoptosis along with the expression of B-cell lymphoma-2 (Bcl2) and BCL2-associated X (Bax) was explored. After that, RNA-seq and bioinformatics analysis revealed the lncRNAs-miRNAs-mRNAs ceRNA network related to hypoxia adaptation of H9c2 cells modulated by LRMA. The final step was to detect the expression of cyclin T2 (CCNT2), cyclin-dependent kinase 9 (CDK9), and forkhead box P1 (FOXP1). The results revealed that LRMA significantly attenuated the reduction of viability and protected H9c2 cells under hypoxic condition. In addition, LRMA not only restored the proliferation activity of H9c2 cells but also promoted apoptosis through stimulating Bcl2 enhancement and Bax inhibition induced by hypoxic stimulation. This study also identified distinct pairs of ceRNA and the hypoxia-related differentially expressed genes regulated by LRMA, including 862 lncRNAs (694 upregulated and 168 downregulated), 7 miRNAs (1 upregulated and 6 downregulated), and 351 mRNAs (226 upregulated and 125 downregulated). The results also suggested that the hypoxia-related genes regulated by LRMA were mainly enriched in cell proliferation and division, cell energy metabolism, inflammatory response, and cell development. However, LRMA could effectively promote the proliferation of H9c2 cells through enhancing CCNT2, CDK9, and FOXP1 under hypoxic conditions. Overall, LRMA could alleviate the proliferation inhibition and apoptosis in hypoxia-induced H9c2 cells.

show abstract

Section: Discussionsupporting

confidence: 93%

The Effect of Lycium ruthenicum Murry Anthocyanins on the Apoptosis and Proliferation of H9c2 Cell Induced by Hypoxia

Li,

Wang,

et al. 2023

Journal of Food Biochemistry

View full text Add to dashboard Cite

show abstract

“…Matrix factorization has the characteristics of interpretability and dimensionality reduction, so it can be combined with machine learning and deep learning models and applied in various fields [15,16]. In computer vision, image classification uses matrix factorization to extract image features, which can more accurately identify different objects in the image.…”

Section: Matrix Factorizationmentioning

confidence: 99%

GCN-MF: A graph convolutional network based on matrix factorization for recommendation

Junxi,

Wang,

Chen

2024

Innov. technol. adv.

View full text Add to dashboard Cite

With the increasing development of information technology and the rise of big data, the Internet has entered the era of information overload. While users enjoy the convenience brought by big data to their daily lives, they also face more and more information filtering and selection problems. In this context, recommendation systems have emerged, and existing recommendation systems cannot effectively deal with the problem of data sparsity. Therefore, this paper proposes a graph convolutional network based on matrix factorization for recommendation. The embedding layer uses matrix factorization instead of neighborhood aggregation, and the interaction layer uses multi-layer neural networks instead of simple inner products. Finally, on the Movielens-1M, Yelp and Gowalla public data set, NDCG and Recall are better than the existing baseline model, which effectively alleviates the data sparsity problem.

show abstract

Clustering by sparse orthogonal NMF and interpretable neural network

Gai,

Liu

2023

Multimedia Systems

View full text Add to dashboard Cite

Learning matrix factorization with scalable distance metric and regularizer

Cited by 7 publications

References 33 publications

The Effect of Lycium ruthenicum Murry Anthocyanins on the Apoptosis and Proliferation of H9c2 Cell Induced by Hypoxia

The Effect of Lycium ruthenicum Murry Anthocyanins on the Apoptosis and Proliferation of H9c2 Cell Induced by Hypoxia

GCN-MF: A graph convolutional network based on matrix factorization for recommendation

Clustering by sparse orthogonal NMF and interpretable neural network

Contact Info

Product

Resources

About