Ordinal Mixed Membership Models

Seppo Virtanen, Mark Girolami
Proceedings of the 32nd International Conference on Machine Learning, PMLR 37:588-596, 2015.

Abstract

We present a novel class of mixed membership models for joint distributions of groups of observations that co-occur with ordinal response variables for each group for learning statistical associations between the ordinal response variables and the observation groups. The class of proposed models addresses a requirement for predictive and diagnostic methods in a wide range of practical contemporary applications. In this work, by way of illustration, we apply the models to a collection of consumer-generated reviews of mobile software applications, where each review contains unstructured text data accompanied with an ordinal rating, and demonstrate that the models infer useful and meaningful recurring patterns of consumer feedback. We also compare the developed models to relevant existing works, which rely on improper statistical assumptions for ordinal variables, showing significant improvements both in predictive ability and knowledge extraction.

Cite this Paper

BibTeX
@InProceedings{pmlr-v37-virtanen15, title = {Ordinal Mixed Membership Models}, author = {Virtanen, Seppo and Girolami, Mark}, booktitle = {Proceedings of the 32nd International Conference on Machine Learning}, pages = {588--596}, year = {2015}, editor = {Bach, Francis and Blei, David}, volume = {37}, series = {Proceedings of Machine Learning Research}, address = {Lille, France}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v37/virtanen15.pdf}, url = {https://proceedings.mlr.press/v37/virtanen15.html}, abstract = {We present a novel class of mixed membership models for joint distributions of groups of observations that co-occur with ordinal response variables for each group for learning statistical associations between the ordinal response variables and the observation groups. The class of proposed models addresses a requirement for predictive and diagnostic methods in a wide range of practical contemporary applications. In this work, by way of illustration, we apply the models to a collection of consumer-generated reviews of mobile software applications, where each review contains unstructured text data accompanied with an ordinal rating, and demonstrate that the models infer useful and meaningful recurring patterns of consumer feedback. We also compare the developed models to relevant existing works, which rely on improper statistical assumptions for ordinal variables, showing significant improvements both in predictive ability and knowledge extraction.} }
Endnote
%0 Conference Paper %T Ordinal Mixed Membership Models %A Seppo Virtanen %A Mark Girolami %B Proceedings of the 32nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2015 %E Francis Bach %E David Blei %F pmlr-v37-virtanen15 %I PMLR %P 588--596 %U https://proceedings.mlr.press/v37/virtanen15.html %V 37 %X We present a novel class of mixed membership models for joint distributions of groups of observations that co-occur with ordinal response variables for each group for learning statistical associations between the ordinal response variables and the observation groups. The class of proposed models addresses a requirement for predictive and diagnostic methods in a wide range of practical contemporary applications. In this work, by way of illustration, we apply the models to a collection of consumer-generated reviews of mobile software applications, where each review contains unstructured text data accompanied with an ordinal rating, and demonstrate that the models infer useful and meaningful recurring patterns of consumer feedback. We also compare the developed models to relevant existing works, which rely on improper statistical assumptions for ordinal variables, showing significant improvements both in predictive ability and knowledge extraction.
RIS
TY - CPAPER TI - Ordinal Mixed Membership Models AU - Seppo Virtanen AU - Mark Girolami BT - Proceedings of the 32nd International Conference on Machine Learning DA - 2015/06/01 ED - Francis Bach ED - David Blei ID - pmlr-v37-virtanen15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 37 SP - 588 EP - 596 L1 - http://proceedings.mlr.press/v37/virtanen15.pdf UR - https://proceedings.mlr.press/v37/virtanen15.html AB - We present a novel class of mixed membership models for joint distributions of groups of observations that co-occur with ordinal response variables for each group for learning statistical associations between the ordinal response variables and the observation groups. The class of proposed models addresses a requirement for predictive and diagnostic methods in a wide range of practical contemporary applications. In this work, by way of illustration, we apply the models to a collection of consumer-generated reviews of mobile software applications, where each review contains unstructured text data accompanied with an ordinal rating, and demonstrate that the models infer useful and meaningful recurring patterns of consumer feedback. We also compare the developed models to relevant existing works, which rely on improper statistical assumptions for ordinal variables, showing significant improvements both in predictive ability and knowledge extraction. ER -
APA
Virtanen, S. & Girolami, M.. (2015). Ordinal Mixed Membership Models. Proceedings of the 32nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 37:588-596 Available from https://proceedings.mlr.press/v37/virtanen15.html.

Related Material