Please use this identifier to cite or link to this item:
Title: Phonetic-enriched text representation for Chinese sentiment analysis with reinforcement learning
Authors: Peng, Haiyun
Ma, Yukun
Poria, Soujanya
Li, Yang
Cambria, Erik
Keywords: Engineering::Computer science and engineering
Issue Date: 2021
Source: Peng, H., Ma, Y., Poria, S., Li, Y. & Cambria, E. (2021). Phonetic-enriched text representation for Chinese sentiment analysis with reinforcement learning. Information Fusion, 70, 88-99.
Journal: Information Fusion
Abstract: The Chinese pronunciation system offers two characteristics that distinguish it from other languages: deep phonemic orthography and intonation variations. In this paper, we hypothesize that these two important properties can play a major role in Chinese sentiment analysis. In particular, we propose two effective features to encode phonetic information and, hence, fuse it with textual information. With this hypothesis, we propose Disambiguate Intonation for Sentiment Analysis (DISA), a network that we develop based on the principles of reinforcement learning. DISA disambiguates intonations for each Chinese character (pinyin) and, hence, learns precise phonetic representations. We also fuse phonetic features with textual and visual features to further improve performance. Experimental results on five different Chinese sentiment analysis datasets show that the inclusion of phonetic features significantly and consistently improves the performance of textual and visual representations and surpasses the state-of-the-art Chinese character-level representations.
ISSN: 1566-2535
DOI: 10.1016/j.inffus.2021.01.005
Schools: School of Computer Science and Engineering 
Rights: © 2021 Elsevier B.V. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:SCSE Journal Articles

Citations 20

Updated on Jun 4, 2023

Web of ScienceTM
Citations 10

Updated on Jun 4, 2023

Page view(s)

Updated on Jun 6, 2023

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.