Abstract | ||
---|---|---|
In this work, we propose a novel nonparametric Bayesian method for clustering of data with spatial interdependencies. Specifically, we devise a novel normalized Gamma process, regulated by a simplified (pointwise) Markov random field (Gibbsian) distribution with a countably infinite number of states. As a result of its construction, the proposed model allows for introducing spatial dependencies in the clustering mechanics of the normalized Gamma process, thus yielding a novel nonparametric Bayesian method for spatial data clustering. We derive an efficient truncated variational Bayesian algorithm for model inference. We examine the efficacy of our approach by considering an image segmentation application using a real-world dataset. We show that our approach outperforms related methods from the field of Bayesian nonparametrics, including the infinite hidden Markov random field model, and the Dirichlet process prior. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1016/j.eswa.2012.05.097 | Expert Syst. Appl. |
Keywords | Field | DocType |
hidden markov random field,clustering mechanic,novel nonparametric bayesian method,markov random field,bayesian nonparametrics,spatially-constrained normalized gamma process,model inference,gamma process,dirichlet process,normalized gamma process,efficient truncated variational bayesian,artificial intelligence,clustering,computer science | Spatial analysis,Dirichlet process,Pattern recognition,Hidden Markov random field,Markov random field,Computer science,Gamma process,Image segmentation,Artificial intelligence,Cluster analysis,Pointwise | Journal |
Volume | Issue | ISSN |
39 | 17 | 0957-4174 |
Citations | PageRank | References |
1 | 0.35 | 16 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sotirios P. Chatzis | 1 | 250 | 24.25 |
Dimitrios Korkinof | 2 | 28 | 3.68 |
Yiannis Demiris | 3 | 938 | 86.45 |