Lexicon information in neural sentiment analysis: a multi-task learning approach

Code

• github.com/ltgoslo/mtl_lex
  OfficialIn paperpytorch★ 0
• github.com/ltgoslo/norsentlex
  OfficialIn papernone★ 0

Abstract

This paper explores the use of multi-task learning (MTL) for incorporating external knowledge in neural models. Specifically, we show how MTL can enable a BiLSTM sentiment classifier to incorporate information from sentiment lexicons. Our MTL set-up is shown to improve model performance (compared to a single-task set-up) on both English and Norwegian sentence-level sentiment datasets. The paper also introduces a new sentiment lexicon for Norwegian.

Tasks

Reproductions