Learning document representations using subspace multinomial model

Document representation, subspace modelling, topic identification, latent topic discovery

Subspace multinomial model (SMM) is a log-linear model and can be used for learning low dimensional continuous representation for discrete data. SMMand its variants have been used for speaker verification based on prosodic features and phonotactic language recognition. In this paper, we propose a new variant of SMM that introduces sparsity and call the resulting model as `1 SMM. We show that `1 SMM can be used for learning document representations that are helpful in topic identification or classification and clustering tasks. Our experiments in document classification show that SMM achieves comparable results to models such as latent Dirichlet allocation and sparse topical coding, while having a useful property that the resulting document vectors are Gaussian distributed.

Subspace multinomial model (SMM) is a log-linear model and can be used for learning low dimensional continuous representation for discrete data. SMMand its variants have been used for speaker verification based on prosodic features and phonotactic language recognition. In this paper, we propose a new variant of SMM that introduces sparsity and call the resulting model as `1 SMM. We show that `1 SMM can be used for learning document representations that are helpful in topic identification or classification and clustering tasks. Our experiments in document classification show that SMM achieves comparable results to models such as latent Dirichlet allocation and sparse topical coding, while having a useful property that the resulting document vectors are Gaussian distributed.