Mixing Dirichlet Topic Models and Word Embeddings to Make lda2vec

TL;DR

Lda2vec combines word embeddings and topic models to produce interpretable, sparse document representations while learning dense word vectors, enhancing semantic understanding in language processing.

Contribution

This work introduces lda2vec, a novel model that jointly learns word vectors and topic distributions with a simple, differentiable framework for interpretable document representations.

Findings

Produces sparse, interpretable document mixtures

Jointly learns word vectors and topic relationships

Easily integrated into existing frameworks

Abstract

Distributed dense word vectors have been shown to be effective at capturing token-level semantic and syntactic regularities in language, while topic models can form interpretable representations over documents. In this work, we describe lda2vec, a model that learns dense word vectors jointly with Dirichlet-distributed latent document-level mixtures of topic vectors. In contrast to continuous dense document representations, this formulation produces sparse, interpretable document mixtures through a non-negative simplex constraint. Our method is simple to incorporate into existing automatic differentiation frameworks and allows for unsupervised document representations geared for use by scientists while simultaneously learning word vectors and the linear relationships between them.