Article
Computer Science, Artificial Intelligence
Ping Deng, Tianrui Li, Hongjun Wang, Shi-Jinn Horng, Zeng Yu, Xiaomin Wang
Summary: The objective of co-clustering is to identify similarity blocks between the sample set and feature set simultaneously. The nonnegative matrix tri-factorization algorithm is an important tool for co-clustering. To address the impact of noise, a tri-regularized NMTF model was proposed to optimize the performance and generalization ability of the model effectively.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Diego Salazar, Juan Rios, Sara Aceros, Oscar Florez-Vargas, Carlos Valencia
Summary: The integration of data from different sources using NMF and jNMF methods can facilitate clustering and interpretation, but they may not effectively identify nonlinear patterns. A new variant called Kernel jNMF is proposed to address this limitation, showing better performance in clustering and interpretation.
Article
Genetics & Heredity
Bingbo Wang, Xiujuan Ma, Minghui Xie, Yue Wu, Yajun Wang, Ran Duan, Chenxing Zhang, Liang Yu, Xingli Guo, Lin Gao
Summary: This paper introduces CBP-JMF, a practical tool for discovering CBPs based on a joint non-negative matrix tri-factorization framework. Applied to identifying four subtypes of breast cancer, CBP-JMF shows significant effectiveness in interpreting disease subtypes.
FRONTIERS IN GENETICS
(2021)
Article
Computer Science, Information Systems
Jin Deng, Weiming Zeng, Sizhe Luo, Wei Kong, Yuhu Shi, Ying Li, Hua Zhang
Summary: The study introduces a novel method, MDJNMF, for identifying biologically functional modules associated with sarcoma lung metastasis by integrating histopathology images and genomic data, which could potentially reveal diagnostic biomarkers.
INFORMATION SCIENCES
(2021)
Article
Engineering, Civil
Haicheng Qu, Jiangtao Guo, Yanji Jiang
Summary: In this study, we propose a recommendation algorithm called LG-DropEdge to mitigate overfitting issues in deep neural network-based recommendation algorithms. By combining lightweight graph convolutional networks and DropEdge techniques, the proposed algorithm can improve the precision of recommendation results by aggregating embedding results at different layers. Experimental results on multiple datasets demonstrate its effectiveness.
JOURNAL OF ADVANCED TRANSPORTATION
(2022)
Article
Green & Sustainable Science & Technology
Md Sanowar Hossain, Ripon K. Chakrabortty, Sondoss El Sawah, Michael J. Ryan
Summary: Traditional product architecture design often neglects interface complexity and conflicting goals, resulting in reduced design robustness and increased complexity for product assembly and recovery. To address these issues, this paper proposes a sustainable modular product architecture that enhances assemblability and recovery. By considering sustainable modular drivers along with architectural drivers, a bi-level leader-follower joint optimization model is formulated to optimize module granularity and reduce assembly costs within a coherent framework.
JOURNAL OF CLEANER PRODUCTION
(2021)
Article
Biochemical Research Methods
Qi Dang, Yong Liang, Dong Ouyang, Rui Miao, Caijin Ling, Xiaoying Liu, Shengli Xie
Summary: Drug repositioning (DR) is a strategy to find new targets for existing drugs, which plays an important role in reducing the costs, time, and risk of traditional drug development. In this study, a self-paced non-negative matrix tri-factorization (SPLNMTF) model is proposed to integrate different biological data and improve the learning ability of the model.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Computer Science, Information Systems
Shiming He, Meng Guo, Zhuozhou Li, Ying Lei, Siyuan Zhou, Kun Xie, Neal N. Xiong
Summary: KPI clustering is important for AIOps when dealing with a large number of KPIs. This approach divides KPIs into classes and applies the same model to detect anomalies or predict outcomes for the KPIs in each class, reducing computational overhead. However, irregular KPIs caused by varying sampling strategies have not been fully addressed. This study proposes an iterative clustering scheme based on matrix factorization to solve the problem of clustering irregular KPIs and achieves higher NMI compared to non-iterative clustering.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Interdisciplinary Applications
Jun Lang, Chongyang Lin
Summary: Nuclear magnetic resonance (NMR) spectroscopy is widely used in chemistry and medicine to study matter composition and protein spatial structure. To speed up NMR signal acquisition, Non-Uniform Sampling (NUS) methods and mathematical algorithms are used to recover the original NMR signals from NUS data. This paper proposes a Fast Tri-Factorization (FTF) method that decomposes the low-rank Hankel matrix into three small-scale matrices, reducing the computational complexity of singular value decomposition (SVD) and speeding up NMR reconstruction.
JOURNAL OF COMPUTATIONAL SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Xiaoyi Jia, Xiaoyu Dong, Meng Chen, Xiaohui Yu
Summary: This paper proposes a novel Imputation Model for traffic Congestion data, CIM, based on joint matrix factorization to estimate missing congestion values. Experimental results show that modeling the periodicity, road similarity, and temporal coherence of congestion patterns simultaneously is effective.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Biochemical Research Methods
Pietro Pinoli, Gaia Ceddia, Stefano Ceri, Marco Masseroli
Summary: In this study, a Non-negative Matrix Tri-Factorization based approach is proposed to predict synergistic drug pairs in specific cell lines by integrating different data types. The results show that our method achieves high performance when cell line genomic data and rich data about synergistic drugs are considered.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Jun Wang, Xi Chen, Zhengtian Wu, Maozu Guo, Guoxian Yu
Summary: CDPMiner is a method for discovering cooperative driver pathways through multiplex network embedding, which can model the relational and attribute information of multi-type molecules. It optimizes the relations between genes and pathways and constructs an attributed multiplex network, and then uses deep joint matrix factorization to mine essential information for pathway-level analysis and reconstruct the pathway interaction network. CDPMiner leverages the reconstructed network and mutation data to define the driver weight between pathways and discover cooperative driver pathways.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Computer Science, Information Systems
Juncheng Hu, Yongheng Xing, Mo Han, Feng Wang, Kuo Zhao, Xilong Che
Summary: Cluster analysis in Heterogeneous Information Networks (HINs) is an effective method that preserves semantic information and structure information. This study applied Nonnegative Matrix Tri-Factorization (NMTF) for cluster analysis and achieved better clustering results than existing algorithms.
TSINGHUA SCIENCE AND TECHNOLOGY
(2022)
Article
Genetics & Heredity
Zhihong Zhang, Meiping Jiang, Dongjie Wu, Wang Zhang, Wei Yan, Xilong Qu
Summary: The NTMEP model proposed in this study utilizes non-negative matrix tri-factorization technology to improve the condition of protein-protein interaction networks for better prediction of essential proteins. The experiments conducted on publicly available datasets demonstrate the superior performance of the NTMEP model compared to existing methods in predicting essential proteins. This finding offers a novel perspective for other applications based on protein-protein interaction networks.
FRONTIERS IN GENETICS
(2021)
Article
Biochemical Research Methods
Aditya Gorla, Brandon Jew, Luke Zhang, Jae Hoon Sul
Summary: The researchers introduced xGAP, an extensible Genome Analysis Pipeline, for analyzing DNA-seq data with high automation, scalability, and accuracy. The pipeline implements modified GATK best practices, allowing efficient parallel processing of data and achieving consistent, high-quality results across various environments.