Improving sentiment analysis accuracy with emoji embedding

Due to the diversity and variability of Chinese syntax and semantics, accurately identifying and distinguishing individual emotions from online texts is challenging. To overcome this limitation, we incorporate a new source of individual sentiment, emojis, which contain thousands of graphic symbols a...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Chuchu Liu, Fan Fang, Xu Lin, Tie Cai, Xu Tan, Jianguo Liu, Xin Lu
Formato: article
Lenguaje:EN
Publicado: KeAi Communications Co., Ltd. 2021
Materias:
Acceso en línea:https://doaj.org/article/38a347918b224dd0a226ef038b10de03
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:38a347918b224dd0a226ef038b10de03
record_format dspace
spelling oai:doaj.org-article:38a347918b224dd0a226ef038b10de032021-11-10T04:41:44ZImproving sentiment analysis accuracy with emoji embedding2666-449610.1016/j.jnlssr.2021.10.003https://doaj.org/article/38a347918b224dd0a226ef038b10de032021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S2666449621000529https://doaj.org/toc/2666-4496Due to the diversity and variability of Chinese syntax and semantics, accurately identifying and distinguishing individual emotions from online texts is challenging. To overcome this limitation, we incorporate a new source of individual sentiment, emojis, which contain thousands of graphic symbols and are increasingly being used for expressing emotion in online conversations. We examined popular sentiment analysis algorithms, including rule-based and classification algorithms, to evaluate the impact of supplementing emojis as additional features to improve the algorithm performance. Emojis were also translated into corresponding sentiment words when constructing features for comparison with those directly generated from emoji label words. In addition, considering different functions of emojis in texts, we classified all posts in the dataset by their emoji usage and examined the changes in algorithm performance. We found that emojis are effective as expanding features for improving the accuracy of sentiment analysis algorithms, and the algorithm performance can be further increased by taking different emoji usages into consideration. In this study, we developed an improved emoji-embedding model based on Bi-LSTM (namely, CEmo-LSTM), which achieves the highest accuracy (around 0.95) when analyzing online Chinese texts. We applied the CEmo-LSTM algorithm to a large dataset collected from Weibo from December 1, 2019 to March 20, 2020 to understand the sentiment evolution of online users during the COVID-19 pandemic. We found that the pandemic remarkably impacted individual sentiments and caused more passive emotions (e.g., horror and sadness). Our novel emoji-embedding algorithm creatively combined emojis as well as emoji usage with the sentiment analysis model and can handle emotion mining tasks more effectively and efficiently.Chuchu LiuFan FangXu LinTie CaiXu TanJianguo LiuXin LuKeAi Communications Co., Ltd.articleSentiment analysisEmojiCEmo-LSTMSentiment evolutionCOVID-19Risk in industry. Risk managementHD61ENJournal of Safety Science and Resilience, Vol 2, Iss 4, Pp 246-252 (2021)
institution DOAJ
collection DOAJ
language EN
topic Sentiment analysis
Emoji
CEmo-LSTM
Sentiment evolution
COVID-19
Risk in industry. Risk management
HD61
spellingShingle Sentiment analysis
Emoji
CEmo-LSTM
Sentiment evolution
COVID-19
Risk in industry. Risk management
HD61
Chuchu Liu
Fan Fang
Xu Lin
Tie Cai
Xu Tan
Jianguo Liu
Xin Lu
Improving sentiment analysis accuracy with emoji embedding
description Due to the diversity and variability of Chinese syntax and semantics, accurately identifying and distinguishing individual emotions from online texts is challenging. To overcome this limitation, we incorporate a new source of individual sentiment, emojis, which contain thousands of graphic symbols and are increasingly being used for expressing emotion in online conversations. We examined popular sentiment analysis algorithms, including rule-based and classification algorithms, to evaluate the impact of supplementing emojis as additional features to improve the algorithm performance. Emojis were also translated into corresponding sentiment words when constructing features for comparison with those directly generated from emoji label words. In addition, considering different functions of emojis in texts, we classified all posts in the dataset by their emoji usage and examined the changes in algorithm performance. We found that emojis are effective as expanding features for improving the accuracy of sentiment analysis algorithms, and the algorithm performance can be further increased by taking different emoji usages into consideration. In this study, we developed an improved emoji-embedding model based on Bi-LSTM (namely, CEmo-LSTM), which achieves the highest accuracy (around 0.95) when analyzing online Chinese texts. We applied the CEmo-LSTM algorithm to a large dataset collected from Weibo from December 1, 2019 to March 20, 2020 to understand the sentiment evolution of online users during the COVID-19 pandemic. We found that the pandemic remarkably impacted individual sentiments and caused more passive emotions (e.g., horror and sadness). Our novel emoji-embedding algorithm creatively combined emojis as well as emoji usage with the sentiment analysis model and can handle emotion mining tasks more effectively and efficiently.
format article
author Chuchu Liu
Fan Fang
Xu Lin
Tie Cai
Xu Tan
Jianguo Liu
Xin Lu
author_facet Chuchu Liu
Fan Fang
Xu Lin
Tie Cai
Xu Tan
Jianguo Liu
Xin Lu
author_sort Chuchu Liu
title Improving sentiment analysis accuracy with emoji embedding
title_short Improving sentiment analysis accuracy with emoji embedding
title_full Improving sentiment analysis accuracy with emoji embedding
title_fullStr Improving sentiment analysis accuracy with emoji embedding
title_full_unstemmed Improving sentiment analysis accuracy with emoji embedding
title_sort improving sentiment analysis accuracy with emoji embedding
publisher KeAi Communications Co., Ltd.
publishDate 2021
url https://doaj.org/article/38a347918b224dd0a226ef038b10de03
work_keys_str_mv AT chuchuliu improvingsentimentanalysisaccuracywithemojiembedding
AT fanfang improvingsentimentanalysisaccuracywithemojiembedding
AT xulin improvingsentimentanalysisaccuracywithemojiembedding
AT tiecai improvingsentimentanalysisaccuracywithemojiembedding
AT xutan improvingsentimentanalysisaccuracywithemojiembedding
AT jianguoliu improvingsentimentanalysisaccuracywithemojiembedding
AT xinlu improvingsentimentanalysisaccuracywithemojiembedding
_version_ 1718440543329976320