Building Language Models for Text with Named Entities

  • 2018-05-13 07:46:12
  • Md Rizwan Parvez, Saikat Chakraborty, Baishakhi Ray, Kai-Wei Chang
  • 3

Abstract

Text in many domains involves a significant amount of named entities.Predict- ing the entity names is often challenging for a language model as theyappear less frequent on the training corpus. In this paper, we propose a noveland effective approach to building a discriminative language model which canlearn the entity names by leveraging their entity type information. We alsointroduce two benchmark datasets based on recipes and Java programming codes,on which we evalu- ate the proposed model. Experimental re- sults show that ourmodel achieves 52.2% better perplexity in recipe generation and 22.06% on codegeneration than the state-of-the-art language models.

 

Quick Read (beta)

loading the full paper ...