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 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 assessment