Fusion Local and Global Aspect-based Sentiment Analysis

EasyChair Preprint 9977

Abstract

Aspect-based sentiment classification is an important task in natural language processing research, and in response to the fact that most studies at this stage ignore the influence of contextual semantic information on the sentiment polarity of aspect words, the BiLSTM-LCF model proposed in this paper combines local aspect word feature extraction and global contextual semantic information extraction based on Bi-directional Long Short-Term Memory (BiLSTM), and after a multi-headed attention mechanism to enhance the local aspect word sentiment representation. Comparative experiments were conducted on the restaurant and laptop datasets of the SEMEVAL2014 evaluation task. The experimental results show that the model proposed in this paper achieves good classification results in the aspect-level sentiment analysis task of text reviews. The method provides a new idea for ABSA task development.Aspect-based sentiment classification is an important task in natural language processing research, and in response to the fact that most studies at this stage ignore the influence of contextual semantic information on the sentiment polarity of aspect words, the BiLSTM-LCF model proposed in this paper combines local aspect word feature extraction and global contextual semantic information extraction based on Bi-directional Long Short-Term Memory (BiLSTM), and after a multi-headed attention mechanism to enhance the local aspect word sentiment representation.

Keyphrases: BERT, Interactive attention, NLP, Sentiment Analysis

@booklet{EasyChair:9977,
  author    = {Yusheng Zhou and Lei Li and Jie Li},
  title     = {Fusion Local and Global Aspect-based Sentiment Analysis},
  howpublished = {EasyChair Preprint 9977},
  year      = {EasyChair, 2023}}