Probabilistic Tensor Analysis with Akaike and Bayesian Information Criteria
Conference Proceeding Article
From data mining to computer vision, from visual surveillance to biometrics research, from biomedical imaging to bioinformatics, and from multimedia retrieval to information management, a large amount of data are naturally represented by multidimensional arrays, i.e., tensors. However, conventional probabilistic graphical models with probabilistic inference only model data in vector format, although they are very important in many statistical problems, e.g., model selection. Is it possible to construct multilinear probabilistic graphical models for tensor format data to conduct probabilistic inference, e.g., model selection? This paper provides a positive answer based on the proposed decoupled probabilistic model by developing the probabilistic tensor analysis (PTA), which selects suitable model for tensor format data modeling based on Akaike information criterion (AIC) and Bayesian information criterion (BIC). Empirical studies demonstrate that PTA associated with AIC and BIC selects correct number of models.
Databases and Information Systems | Numerical Analysis and Scientific Computing
Data Management and Analytics
Neural Information Processing: 14th International Conference, ICONIP 2007, Kitakyushu, Japan, 13-16 November: Revised Selected Papers, Part I
City or Country
TAO, Dacheng; SUN, Jimeng; SHEN, Jialie; Wu, Xindong; LI, Xuelong; Maybank, Stephen J.; and Faloutsos, Christos.
Probabilistic Tensor Analysis with Akaike and Bayesian Information Criteria. (2007). Neural Information Processing: 14th International Conference, ICONIP 2007, Kitakyushu, Japan, 13-16 November: Revised Selected Papers, Part I. 4984, 791-801. Research Collection School Of Information Systems.
Available at: http://ink.library.smu.edu.sg/sis_research/399