Neural variational sparse topic model for sparse explainable text representation

作者：

Highlights：

• We propose a SR-NSTM for sparse and explainable text representation.

• We extend SR-NSTM to supervised learning tasks with the max-margin posterior constraints.

• Experimental results demonstrate the superiority of our models in perplexity, topic coherence and text classification accuracy.

摘要

•We propose a SR-NSTM for sparse and explainable text representation.•We extend SR-NSTM to supervised learning tasks with the max-margin posterior constraints.•Experimental results demonstrate the superiority of our models in perplexity, topic coherence and text classification accuracy.

论文关键词：Neural sparse topic model,Neural variational inference,Explainable text representation

论文评审过程：Received 3 December 2020, Revised 7 March 2021, Accepted 22 April 2021, Available online 6 May 2021, Version of Record 6 May 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102614