Filters








464,140 Hits in 4.1 sec

Link Prediction via Matrix Factorization [chapter]

Aditya Krishna Menon, Charles Elkan
2011 Lecture Notes in Computer Science  
We propose to solve the link prediction problem in graphs using a supervised matrix factorization approach.  ...  Finally, we propose a novel approach to address the class imbalance problem which is common in link prediction by directly optimizing for a ranking loss.  ...  We first make a case for matrix factorization to serve as a foundation for a general purpose link prediction model.  ... 
doi:10.1007/978-3-642-23783-6_28 fatcat:tsxawuytnze3dh5iub4fpln5zm

Neural graph embeddings via matrix factorization for link prediction: smoothing or truncating negatives? [article]

Asan Agibetov
2022 arXiv   pre-print
To circumvent this issue, we propose an improved approach to learning low-rank factorization embeddings that incorporate information from such unlikely pairs of nodes and show that it can improve the link  ...  This design choice has been mostly popularized by the direct application of the highly-successful word embedding algorithm word2vec to predicting the formation of new links in social, co-citation, and  ...  In developing our approach, we ponder whether the matrix factorization of PMI matrix is the best way to obtain node representations suitable for link prediction of real-world networks?  ... 
arXiv:2011.09907v2 fatcat:tbu6w7ilrvd6nehvb6sazr6z7q

Link Prediction via Convex Nonnegative Matrix Factorization on Multiscale Blocks

Enming Dong, Jianping Li, Zheng Xie
2014 Journal of Applied Mathematics  
We use this insight to give a probabilistic latent variable model for finding missing links by convex nonnegative matrix factorization with block detection.  ...  Low rank matrices approximations have been used in link prediction for networks, which are usually global optimal methods and lack of using the local information.  ...  Then different from the low rank matrices approximations algorithms already used for link predictions, we use a new low rank matrices approximations algorithm named convex nonnegative matrix factorization  ... 
doi:10.1155/2014/786156 fatcat:u664sy3jtradlewhbr4hbqeiha

A perturbation-based framework for link prediction via non-negative matrix factorization

Wenjun Wang, Fei Cai, Pengfei Jiao, Lin Pan
2016 Scientific Reports  
doi:10.1038/srep38938 pmid:27976672 pmcid:PMC5156920 fatcat:tl32wl25kbco7hnzaeaabs4rvq

Scalable Temporal Latent Space Inference for Link Prediction in Dynamic Social Networks

Linhong Zhu, Dong Guo, Junming Yin, Greg Ver Steeg, Aram Galstyan
2016 IEEE Transactions on Knowledge and Data Engineering  
We propose a scalable temporal latent space model for link prediction in dynamic social networks, where the goal is to predict links over time based on a sequence of previous graph snapshots.  ...  Empirically, we demonstrate that our model, when evaluated on a number of real-world dynamic networks, significantly outperforms existing approaches for temporal link prediction in terms of both scalability  ...  One of the most widely used methods for inferring low-rank latent space in networks is via matrix factorization with the "block-coordinate gradient descent (BCGD)" algorithm [6] , [38] , which has been  ... 
doi:10.1109/tkde.2016.2591009 fatcat:vtlpkq64azd25etxbfiafuzwle

Link Quality Prediction via a Neighborhood-Based Nonnegative Matrix Factorization Model for Wireless Sensor Networks

Yuxin Zhao, Shenghong Li, Jia Hou
2015 International Journal of Distributed Sensor Networks  
By introducing complex network theory and machine learning techniques, we propose a neighborhood-based nonnegative matrix factorization model to predict link quality in wireless sensor networks.  ...  Link quality prediction is an important approach to solve this problem.  ...  Our proposed neighborhood-based nonnegative matrix factorization model for link quality prediction is described in detail in Section 3.  ... 
doi:10.1155/2015/828493 fatcat:fp5unu7mnbh5xisghjevijvyjq

Cold-Start Link Prediction via Weighted Symmetric Nonnegative Matrix Factorization with Graph Regularization

Minghu Tang, Wei Yu, Xiaoming Li, Xue Chen, Wenjun Wang, Zhen Liu
2022 Computer systems science and engineering  
In this paper, we propose a framework for solving the cold-start link prediction problem, a joint-weighted symmetric nonnegative matrix factorization model fusing graph regularization information, based  ...  Finally, a unified framework for implementing cold-start link prediction is constructed by using a symmetric nonnegative matrix factorization model to integrate the multiple information extracted together  ...  In this paper, considering the advantages of fusing heterogeneous information via a nonnegative matrix factorization (NMF) framework, a cold-start link prediction model, Joint weighted Symmetric NMF integrating  ... 
doi:10.32604/csse.2022.028841 fatcat:ovcxuvpsvvewddwpu4crrfzxnm

A Novel Triple Matrix Factorization Method for Detecting Drug-Side Effect Association Based on Kernel Target Alignment

Xiaoyi Guo, Wei Zhou, Yan Yu, Yijie Ding, Jijun Tang, Fei Guo
2020 BioMed Research International  
To predict potential associations between drugs and side effects, we propose a novel method called the Triple Matrix Factorization- (TMF-) based model.  ...  So, computation-based methods have been developed to accurately and quickly predict side effects.  ...  To predict drug-target interactions, Neighborhood Regularized Logistic Matrix Factorization (NRLMF) [29] , Collaborative Matrix Factorization (CMF) [30] , and Graph Regularized Matrix Factorization (  ... 
doi:10.1155/2020/4675395 pmid:32596314 pmcid:PMC7275954 fatcat:fp2vohkbrnar5jtxs2msxscuoi

Link prediction via latent factor BlockModel

Sheng Gao, Ludovic Denoyer, Patrick Gallinari
2012 Proceedings of the 21st international conference companion on World Wide Web - WWW '12 Companion  
In this paper we address the problem of link prediction in networked data, which appears in many applications such as social network analysis or recommender systems.  ...  Extensive experiments on several real world datasets suggest that our proposed model outperforms the other state of the art approaches for link prediction.  ...  Recently, the latent feature based models have been successfully studied for link prediction task [2] [4], which consider link prediction as a matrix completion problem and employ latent matrix factorization  ... 
doi:10.1145/2187980.2188100 dblp:conf/www/GaoDG12 fatcat:lpmhd7p6kvhkdfsp4ch77psrxm

Multiplicative latent factor models for description and prediction of social networks

Peter D. Hoff
2008 Computational and mathematical organization theory  
This approach allows for the graphical description of a social network via the latent factors of the nodes, and provides a framework for the prediction of missing links in network data.  ...  The model can include known predictor information in the form of a regression term, and can represent additional structure via sender-specific and receiver-specific latent factors.  ...  To evaluate the ability of the multiplicative latent factor model to predict missing links, we performed the following prediction experiment on the international conflict data: 1.  ... 
doi:10.1007/s10588-008-9040-4 fatcat:uuoth4pcfnbknnvccltc7ttgja

Link Prediction via Generalized Coupled Tensor Factorisation [article]

Beyza Ermiş and Evrim Acar and A. Taylan Cemgil
2012 arXiv   pre-print
Numerical experiments demonstrate that joint analysis of data from multiple sources via coupled factorisation improves the link prediction performance and the selection of right loss function and tensor  ...  model is crucial for accurately predicting missing links.  ...  The relation between the location entries j and m in X 1 and X 2 are coupled via a common factor over the users.  ... 
arXiv:1208.6231v1 fatcat:hgdswadgzfdotjz2nxfsoxqdlu

Protein-protein interaction prediction via Collective Matrix Factorization

Qian Xu, Evan Wei Xiang, Qiang Yang
2010 2010 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)  
Thus, state-of-the-art methods for link prediction in these networks often cannot give satisfactory prediction results, especially when some networks are extremely sparse.  ...  Noticing that a network structure can be modeled using a matrix model, in this paper, we introduce the wellknown Collective Matrix Factorization (CMF) technique to 'transfer' usable linkage knowledge from  ...  We first introduce matrix factorization (MF) [25] methods for an individual network. Matrix factorization is increasingly popular in link prediction in many domains, including social networks.  ... 
doi:10.1109/bibm.2010.5706537 dblp:conf/bibm/XuXY10 fatcat:gxfiq6gngvfedglc65p6au6e2e

The Innate Immune Database (IIDB)

Martin Korb, Aistair G Rust, Vesteinn Thorsson, Christophe Battail, Bin Li, Daehee Hwang, Kathleen A Kennedy, Jared C Roach, Carrie M Rosenberger, Mark Gilchrist, Daniel Zak, Carrie Johnson (+4 others)
2008 BMC Immunology  
Our database can be interrogated via a web interface. Genomic annotations and binding site predictions can be automatically viewed with a customized version of the Argo genome browser.  ...  Here, we report the development of a database of computationally predicted transcription factor binding sites and related genomic features for a set of over 2000 murine immune genes of interest.  ...  Tutorial link).  ... 
doi:10.1186/1471-2172-9-7 pmid:18321385 pmcid:PMC2268913 fatcat:thizhgytyvhp5nanbr3ja627e4

POLE: Polarized Embedding for Signed Networks [article]

Zexi Huang, Arlei Silva, Ambuj Singh
2021 arXiv   pre-print
However, existing models are especially ineffective in predicting conflicts (or negative links) among users.  ...  This is due to a strong correlation between link signs and the network structure, where negative links between polarized communities are too sparse to be predicted even by state-of-the-art approaches.  ...  Existing signed link prediction methods are based on feature engineering [5] , matrix factorization [19, 58] , and graph embedding [57] .  ... 
arXiv:2110.09899v2 fatcat:uz4rzmfnd5acrnufl6vm7uw4uy

Link Prediction on Evolving Data Using Matrix and Tensor Factorizations

Evrim Acar, Daniel M. Dunlavy, Tamara G. Kolda
2009 2009 IEEE International Conference on Data Mining Workshops  
Specifically, we look at bipartite graphs changing over time and consider matrix-and tensorbased methods for predicting links.  ...  In this paper, we consider the problem of temporal link prediction: Given link data for time periods 1 through T , can we predict the links in time period T +1?  ...  The goal is to predict the links at time T + 1 by analyzing the link structure of Z. We consider both matrix-and tensor-based methods for link prediction.  ... 
doi:10.1109/icdmw.2009.54 dblp:conf/icdm/AcarDK09 fatcat:4itnhpjh3bfejasc66bybuh7oa
« Previous Showing results 1 — 15 out of 464,140 results