Filters








174,441 Hits in 6.1 sec

A Bayesian Matrix Factorization Model for Relational Data [article]

Ajit P. Singh, Geoffrey Gordon
2012 arXiv   pre-print
We frame a large class of relational learning problems as matrix factorization problems, and propose a hierarchical Bayesian model.  ...  Relational learning can be used to augment one data source with other correlated sources of information, to improve predictive accuracy.  ...  Gharahamani for their insightful comments. The authors gratefully acknowledge the support of NSF, under SBE-0836012.  ... 
arXiv:1203.3517v1 fatcat:sclxvlnmzfeffihvj53shbagci

A Bayesian Matrix Factorization Model for Relational Data

Ajit P. Singh, Geoffrey J. Gordon
2018
We frame a large class of relational learning problems as matrix factorization problems, and propose a hierarchical Bayesian model.  ...  Relational learning can be used to augment one data source with other correlated sources of information, to improve predictive accuracy.  ...  Gharahamani for their insightful comments. The authors gratefully acknowledge the support of NSF, under SBE-0836012.  ... 
doi:10.1184/r1/6475406 fatcat:r6hycl7ncvcjpcvds46vvkzsme

Patent Keyword Analysis of Disaster Artificial Intelligence Using Bayesian Network Modeling and Factor Analysis

Sangsung Park, Sunghae Jun
2020 Sustainability  
In this paper, we analyze the patent documents related to disaster AI technology. We propose Bayesian network modeling and factor analysis for the technology analysis of disaster AI.  ...  In order to show how the proposed model can be applied to a real problem, we carried out a case study to collect and analyze the patent data related to disaster AI.  ...  So, we made a matrix with 16,875 rows (patents) and 251,358 columns (terms) as a structured set of data for Bayesian network modeling and factor analysis.  ... 
doi:10.3390/su12020505 fatcat:huvf3cwrfrap5ob5dpsvxuweia

Bayesian Factor Analysis When Only a Sample Covariance Matrix Is Available

Kentaro Hayashi, Marina Arav
2006 Educational and Psychological Measurement  
We propose a simple method for computing the posterior estimates of Bayesian factor analysis using only the sample variance-covariance matrix without the entire data set.  ...  Bayesian Factor Analysis Press and Shigemasu (1989, 1997) proposed a new type of BFA model.  ... 
doi:10.1177/0013164405278583 fatcat:rxxh7754grcftpuwpgwgxeqf2q

A Discovery of Physiological and Psychological Connection Based on Bayesian Network

Ali ABRISHAMIAN, Kentaro FUKUTA, Junichiro WAKATSUKI, Takashi UOZUMI
2012 Transactions of Japan Society of Kansei Engineering  
Furthermore, special appreciation is extended to CBS Inc. for providing the facial packs used in this research.  ...  Bayesian networks are useful for simulating Kansei information related to psychological and physiological factors.  ...  The Correlated Bayesian network helps us to infer a relation between a material concept (skin 'factors') and human feelings.  ... 
doi:10.5057/jjske.11.167 fatcat:p3ygxf4honhkhewdpg72vepwca

Hierarchical variational Bayesian matrix co-factorization

Jiho Yoo, Seungjin Choi
2012 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
In this paper we present a hierarchical Bayesian model for matrix co-factorization in which we derive a variational inference algorithm to approximately compute posterior distributions over factor matrices  ...  Matrix co-factorization involves jointly decomposing several data matrices to approximate each data matrix as a product of two factor matrices, sharing some factor matrices in the factorization.  ...  on hyperparameters, in order to construct a hierarchical Bayesian model for matrix co-factorization.  ... 
doi:10.1109/icassp.2012.6288275 dblp:conf/icassp/YooC12 fatcat:glgqf4yzd5dltdt3rh5r4y5wbq

The Estimation Process in Bayesian Structural Equation Modeling Approach

Ferra Yanuar
2014 Journal of Physics, Conference Series  
A goodness of fit statistic for assessing the proposed model is presented. An illustrative example with a real data is presented.  ...  A Bayesian approach to SEM may enable models that reflect hypotheses based on complex theory.  ...  Acknowledgments The authors thank to the Institute for Public Health, Ministry of Health, Malaysia, who furnished us the health survey data used in this study.  ... 
doi:10.1088/1742-6596/495/1/012047 fatcat:s4ingkg2wjfrxmewmgm5tszryu

Probabilistic Latent Tensor Factorization Model for Link Pattern Prediction in Multi-relational Networks [article]

Sheng Gao and Ludovic Denoyer and Patrick Gallinari
2012 arXiv   pre-print
To address the issue, we propose a Probabilistic Latent Tensor Factorization (PLTF) model by introducing another latent factor for multiple relation types and furnish the Hierarchical Bayesian treatment  ...  For that, we define the overall relations between object pairs as a link pattern which consists in interaction pattern and connection structure in the network, and then use tensor formalization to jointly  ...  Acknowledgments We would like to thank Kurt T.Miller for providing the Kinship and Countries datasets.  ... 
arXiv:1204.2588v1 fatcat:5utozeoaabetlbhjntdnd6zp4a

Stochastic Relational Models for Large-scale Dyadic Data using MCMC

Shenghuo Zhu, Kai Yu, Yihong Gong
2008 Neural Information Processing Systems  
The latest Bayesian probabilistic matrix factorization [13] reported the state-of-the-art accuracy of matrix factorization on Netflix data.  ...  The models gen- eralize matrix factorization to a supervised learning problem that utilizes attributes of entities in a hierarchical Bayesian framework.  ... 
dblp:conf/nips/ZhuYG08 fatcat:prg6piwyqnazpcbio7v32maqy4

Bayesian Matrix Co-Factorization: Variational Algorithm and Cramér-Rao Bound [chapter]

Jiho Yoo, Seungjin Choi
2011 Lecture Notes in Computer Science  
Matrix factorization is a popular method for collaborative prediction, where unknown ratings are predicted by user and item factor matrices which are determined to approximate a user-item matrix as their  ...  Bayesian matrix factorization is preferred over other methods for collaborative filtering, since Bayesian approach alleviates overfitting, integrating out all model parameters using variational inference  ...  Introduction Matrix factorization is a method for seeking a low-rank latent structure of data, approximating the data matrix as a product of two or more factor matrices.  ... 
doi:10.1007/978-3-642-23808-6_35 fatcat:w6y4uulwvra7nkvbfyl2s2ggli

Bayesian Information Sharing Between Noise And Regression Models Improves Prediction of Weak Effects [article]

Jussi Gillberg, Pekka Marttinen, Matti Pirinen, Antti J Kangas, Pasi Soininen, Marjo-Riitta Järvelin, Mika Ala-Korpela, Samuel Kaski
2013 arXiv   pre-print
infinite factor model as a flexible low-rank noise model.  ...  model structure by introducing a novel Bayesian approach of sharing information between the regression model and the noise model.  ...  ('Bayesian factor regression') and a baseline method of predicting with target data mean.  ... 
arXiv:1310.4362v1 fatcat:lxxpypvdjbcahobbkn5st677oq

Bayesian inference of the number of factors in gene-expression analysis: application to human virus challenge studies

Bo Chen, Minhua Chen, John Paisley, Aimee Zaas, Christopher Woods, Geoffrey S Ginsburg, Alfred Hero, Joseph Lucas, David Dunson, Lawrence Carin
2010 BMC Bioinformatics  
For real data the "true" number of factors is unknown; in our simulations we consider a range of noise variances, and the proposed Bayesian models inferred the number of factors accurately relative to  ...  related non-Bayesian approaches.  ...  For all MCMC results, we employed a singular value decomposition (SVD) of the data matrix to initialize the factor loading and factor score matrix in the FA model, as well as the right-and left-singular  ... 
doi:10.1186/1471-2105-11-552 pmid:21062443 pmcid:PMC3098097 fatcat:44w4l6rtlvbblhhl2vb5pkvgw4

Group-sparse Embeddings in Collective Matrix Factorization [article]

Arto Klami, Guillaume Bouchard, Abhishek Tripathi
2014 arXiv   pre-print
We compare MAP and variational Bayesian solutions based on alternating optimization algorithms and show that the model automatically infers the nature of each factor using group-wise sparsity.  ...  Our approach supports in a principled way continuous, binary and count observations and is efficient for sparse matrices involving missing data.  ...  matrix X m , i.e. the factor k is a private factor for relation m.  ... 
arXiv:1312.5921v2 fatcat:7rzcpupa3jbuvmju4st34q7sd4

Bayesian factor analysis for spatially correlated data: application to cancer incidence data in Scotland

Maura Mezzetti
2011 Statistical Methods & Applications  
A hierarchical Bayesian factor model for multivariate spatially correlated data is proposed.  ...  The proposed model is an extension of a model proposed by Rowe (2003a) and starts from the introduction of separable covariance matrix for the observations.  ...  Here Bayesian factor analysis model for spatially correlated data is proposed.  ... 
doi:10.1007/s10260-011-0177-9 fatcat:p7d5hibhofhu5csiqytuhiad7i

Page 272 of Educational and Psychological Measurement Vol. 66, Issue 2 [page]

2006 Educational and Psychological Measurement  
We propose a simple method for computing the posterior estimates of Bayesian factor analysis using only the sample variance-covariance matrix without the entire data set.  ...  For example, the procedure Proc Factor in the statistical package SAS allows the user to use the sample covariance matrix and the sample cor- relation matrix as its input information (see, e.g., Hatcher  ... 
« Previous Showing results 1 — 15 out of 174,441 results