Improving sentiment analysis accuracy with emoji embedding

Liu, Chuchu; Fang, Fan; Lin, Xu; Tan, Xu; Liu, Jianguo; Lu, Xin

doi:10.1016/j.jnlssr.2021.10.003

Cited by 39 publications

(22 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although there may be slight deviations in the analysis due to the inability to obtain deleted posts, this has no significant impact on the overall conclusions. As different use habits of emojis may lead to different sentiment semantics [ 59 , 60 ], in future work, we tend to further explore the emotional meanings conveyed by emojis by introducing detailed emoji use habits of online users, and analyse the usage patterns of more massive emojis across platforms and contexts.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Emoji use in China: popularity patterns and changes due to COVID-19

et al. 2022

Self Cite

View full text Add to dashboard Cite

Emojis are small pictograms that are frequently embedded within micro-texts to more directly express emotional meanings. To understand the changes in the emoji usage of internet users during the COVID-19 outbreak, we analysed a large dataset collected from Weibo, the most popular Twitter-like social media platform in China, from December 1, 2019, to March 20, 2020. The data contained 38,183,194 microblog posts published by 2,239,472 unique users in Wuhan. We calculated the basic statistics of users’ usage of emojis, topics, and sentiments and analysed the temporal patterns of emoji occurrence. After examining the emoji co-occurrence structure, we finally explored other factors that may affect individual emoji usage. We found that the COVID-19 outbreak greatly changed the pattern of emoji usage; i.e., both the proportion of posts containing emojis and the ratio of users using emojis declined substantially, while the number of posts remained the same. The daily proportion of Happy emojis significantly declined to approximately 32%, but the proportions of Sad- and Encouraging-related emojis rose to 24% and 34%, respectively. Despite a significant decrease in the number of nodes and edges in the emoji co-occurrence network, the average degree of the network increased from 34 to 39.8, indicating that the diversity of emoji usage increased. Most interestingly, we found that male users were more inclined towards using regular textual language with fewer emojis after the pandemic, suggesting that during public crises, male groups appeared to control their emotional display. In summary, the COVID-19 pandemic remarkably impacted individual sentiments, and the normal pattern of emoji usage tends to change significantly following a public emergency. Supplementary Information The online version contains supplementary material available at 10.1007/s10489-022-03195-y.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Emoji use in China: popularity patterns and changes due to COVID-19

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Sentiment analysis is related to text mining which aims to determine public perceptions of a topic of discussion, event or problem. The primary task of sentiment analysis is to classify a particular text and determine whether the emotion of the text or document is positive and negative, or neutral [16]. Each word in the sentence will be given a value, namely one or (1) for a word with a positive sentiment and a negative value for a negative 1 or (-1) for a word with a negative sentiment.…”

Section: Sentiment Analysismentioning

confidence: 99%

Latent Dirichlet Allocation (Lda) Method Analysis About Covid-19 Vaccine on Twitter Social Media

Haay

Setiawan

2022

BAREKENG: J. Il. Mat. & Ter.

View full text Add to dashboard Cite

Twitter is one social media that often provides much information for its users, one of which is information regarding the COVID-19 vaccination. This study aimed to explore and find out what topics are often discussed on Twitter social media. One of which is the topic of COVID-19 vaccination using the Latent Dirichlet Allocation (LDA) method and analysis of the frequency of keywords that often appear with this topic. The Tweet data used in this study was taken from Twitter users worldwide in November 2021. In this study, the results of sentiment analysis were obtained from the tweet data taken, which was divided into positive sentiment and negative sentiment, namely "vaccination" with 40 words and "'Covid19" with 35 words

show abstract

“…The major social network problem is opinion mining or sentiment analysis (SA) 2 . The sentiments, attitudes, ratings, evaluations, feelings, and opinions of people are analyzed 3,4 …”

Section: Introductionmentioning

confidence: 99%

“…2 The sentiments, attitudes, ratings, evaluations, feelings, and opinions of people are analyzed. 3,4 The goal of SA is to determine how the author or another subject feels about specific events or themes, as well as how the speaker feels. Sentiment analysis is performed at the subject, sentence, document, and other levels.…”

Section: Introductionmentioning

confidence: 99%

FBO‐RNN: Fuzzy butterfly optimization‐basedRNN‐LSTMfor extracting sentiments from Twitter Emoji database

Venkataraman

Mohandoss

2023

Concurrency and Computation

View full text Add to dashboard Cite

Social media networks have seen a significant increase in purpose and scale. Sentimental analysis is a component of social networking platforms that uses shared material to infer information about individual sentiments and emotions. In recent years, sentiment analysis (SA) research has grown in popularity. Twitter is the most popular social media platform, with users from many languages and cultures participating. Emojis are used by users to express themselves, and social media platforms contain a wide range of symbols, emotions, and opinions. A novel framework for SA based on Emojis is presented in this article. Initially, the noise-free videos and images are filtered. The dictionary of Jieba was obtained by adding the English Emoji lexicon and English Internet slang lexicon to segment English text. Initially, the Emojis are converted into textual features.Different sentiment classes such as positive, very positive, neutral, negative, and very negative classes are classified using long short-term memory (LSTM) in the recurrent neural network (RNN)-based Fuzzy Butterfly Optimization (FBO) algorithm. The freeware WEKA software tool with different evaluation measures performs experimental investigations. Ultimately, the proposed model demonstrates superior results in the case of SA than other state-of-the-art methods.

show abstract

Improving sentiment analysis accuracy with emoji embedding

Cited by 39 publications

References 16 publications

Emoji use in China: popularity patterns and changes due to COVID-19

Emoji use in China: popularity patterns and changes due to COVID-19

Latent Dirichlet Allocation (Lda) Method Analysis About Covid-19 Vaccine on Twitter Social Media

FBO‐RNN: Fuzzy butterfly optimization‐basedRNN‐LSTMfor extracting sentiments from Twitter Emoji database

Contact Info

Product

Resources

About