Von Mises-Fisher Clustering Models

Siddharth Gopal, Yiming Yang
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(1):154-162, 2014.

Abstract

This paper proposes a suite of models for clustering high-dimensional data on a unit sphere based on Von Mises-Fisher (vMF) distribution and for discovering more intuitive clusters than existing approaches. The proposed models include a) A Bayesian formulation of vMF mixture that enables information sharing among clusters, b) a Hierarchical vMF mixture that provides multi-scale shrinkage and tree structured view of the data and c) a Temporal vMF mixture that captures evolution of clusters in temporal data. For posterior inference, we develop fast variational methods as well as collapsed Gibbs sampling techniques for all three models. Our experiments on six datasets provide strong empirical support in favour of vMF based clustering models over other popular tools such as K-means, Multinomial Mixtures and Latent Dirichlet Allocation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-gopal14, title = {Von Mises-Fisher Clustering Models}, author = {Gopal, Siddharth and Yang, Yiming}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {154--162}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/gopal14.pdf}, url = {https://proceedings.mlr.press/v32/gopal14.html}, abstract = {This paper proposes a suite of models for clustering high-dimensional data on a unit sphere based on Von Mises-Fisher (vMF) distribution and for discovering more intuitive clusters than existing approaches. The proposed models include a) A Bayesian formulation of vMF mixture that enables information sharing among clusters, b) a Hierarchical vMF mixture that provides multi-scale shrinkage and tree structured view of the data and c) a Temporal vMF mixture that captures evolution of clusters in temporal data. For posterior inference, we develop fast variational methods as well as collapsed Gibbs sampling techniques for all three models. Our experiments on six datasets provide strong empirical support in favour of vMF based clustering models over other popular tools such as K-means, Multinomial Mixtures and Latent Dirichlet Allocation.} }
Endnote
%0 Conference Paper %T Von Mises-Fisher Clustering Models %A Siddharth Gopal %A Yiming Yang %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-gopal14 %I PMLR %P 154--162 %U https://proceedings.mlr.press/v32/gopal14.html %V 32 %N 1 %X This paper proposes a suite of models for clustering high-dimensional data on a unit sphere based on Von Mises-Fisher (vMF) distribution and for discovering more intuitive clusters than existing approaches. The proposed models include a) A Bayesian formulation of vMF mixture that enables information sharing among clusters, b) a Hierarchical vMF mixture that provides multi-scale shrinkage and tree structured view of the data and c) a Temporal vMF mixture that captures evolution of clusters in temporal data. For posterior inference, we develop fast variational methods as well as collapsed Gibbs sampling techniques for all three models. Our experiments on six datasets provide strong empirical support in favour of vMF based clustering models over other popular tools such as K-means, Multinomial Mixtures and Latent Dirichlet Allocation.
RIS
TY - CPAPER TI - Von Mises-Fisher Clustering Models AU - Siddharth Gopal AU - Yiming Yang BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/01/27 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-gopal14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 1 SP - 154 EP - 162 L1 - http://proceedings.mlr.press/v32/gopal14.pdf UR - https://proceedings.mlr.press/v32/gopal14.html AB - This paper proposes a suite of models for clustering high-dimensional data on a unit sphere based on Von Mises-Fisher (vMF) distribution and for discovering more intuitive clusters than existing approaches. The proposed models include a) A Bayesian formulation of vMF mixture that enables information sharing among clusters, b) a Hierarchical vMF mixture that provides multi-scale shrinkage and tree structured view of the data and c) a Temporal vMF mixture that captures evolution of clusters in temporal data. For posterior inference, we develop fast variational methods as well as collapsed Gibbs sampling techniques for all three models. Our experiments on six datasets provide strong empirical support in favour of vMF based clustering models over other popular tools such as K-means, Multinomial Mixtures and Latent Dirichlet Allocation. ER -
APA
Gopal, S. & Yang, Y.. (2014). Von Mises-Fisher Clustering Models. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(1):154-162 Available from https://proceedings.mlr.press/v32/gopal14.html.

Related Material