Cumulative restricted Boltzmann machines for ordinal matrix data analysis

Ordinal data is omnipresent in almost all multiuser-generated feedback - questionnaires,preferences etc. This paper investigates modelling of ordinal data with Gaussian restrictedBoltzmann machines (RBMs). In particular, we present the model architecture, learningand inference procedures for both ve...

Full description

Bibliographic Details
Main Authors: Tran, Truyen, Phung, D., Venkatesh, S.
Other Authors: C H Roi
Format: Conference Paper
Published: JMLR 2012
Subjects:
Online Access:http://jmlr.csail.mit.edu/proceedings/
http://hdl.handle.net/20.500.11937/4560
Description
Summary:Ordinal data is omnipresent in almost all multiuser-generated feedback - questionnaires,preferences etc. This paper investigates modelling of ordinal data with Gaussian restrictedBoltzmann machines (RBMs). In particular, we present the model architecture, learningand inference procedures for both vector-variate and matrix-variate ordinal data. We showthat our model is able to capture latent opinion prole of citizens around the world, andis competitive against state-of-art collaborative ltering techniques on large-scale publicdatasets. The model thus has the potential to extend application of RBMs to diversedomains such as recommendation systems, product reviews and expert assessments