Robust simultaneous positive data clustering and unsupervised feature selection using generalized inverted Dirichlet mixture models

作者：

Highlights：

• An algorithm for simultaneous clustering, feature selection and outliers is proposed.

• The proposed approach is based on finite generalized inverted Dirichlet mixture.

• An approach for model selection using minimum message length is developed.

• The model is applied to the challenging problems of visual scenes and objects clustering.

摘要

•An algorithm for simultaneous clustering, feature selection and outliers is proposed.•The proposed approach is based on finite generalized inverted Dirichlet mixture.•An approach for model selection using minimum message length is developed.•The model is applied to the challenging problems of visual scenes and objects clustering.

论文关键词：Positive data,Generalized inverted Dirichlet,Finite mixture,Feature selection,Outliers,Model selection,Images clustering

论文评审过程：Received 16 April 2013, Revised 3 December 2013, Accepted 3 January 2014, Available online 18 January 2014.

论文官网地址：https://doi.org/10.1016/j.knosys.2014.01.007