Semiparametric Latent Topic Modeling on Consumer-Generated Corpora
This work addresses topic modeling challenges for consumer-generated corpora, offering an incremental improvement over existing methods.
The paper tackled the problem of overfitting and sparse topic reconstruction in topic modeling by proposing a semiparametric topic model using nonnegative matrix factorization and semiparametric regression, which improved performance in discovering topic structures for small corpora with limited vocabulary.
Legacy procedures for topic modelling have generally suffered problems of overfitting and a weakness towards reconstructing sparse topic structures. With motivation from a consumer-generated corpora, this paper proposes semiparametric topic model, a two-step approach utilizing nonnegative matrix factorization and semiparametric regression in topic modeling. The model enables the reconstruction of sparse topic structures in the corpus and provides a generative model for predicting topics in new documents entering the corpus. Assuming the presence of auxiliary information related to the topics, this approach exhibits better performance in discovering underlying topic structures in cases where the corpora are small and limited in vocabulary. In an actual consumer feedback corpus, the model also demonstrably provides interpretable and useful topic definitions comparable with those produced by other methods.