Text Analytics on Twitter Text-based Public Sentiment for Covid-19 Vaccine: A Machine Learning Approach

Adamu, Hassan; Jiran, Mat Jasri Bin Mat; Gan, Keng Hoon; Samsudin, Nur-Hana

doi:10.1109/iicaiet51634.2021.9573866

Cited by 5 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, the analysis of tweets on COVID-19 vaccination, according to phase in the vaccine administration, produced insights that were hitherto not available because other researchers (Adamu et al, 2021;Hung et al, 2020;Ong et al, 2022) did not demarcate the tweets according to a timeline. In social media, negative views on COVID-19 vaccination initially overshadowed positive views.…”

Section: Discussionmentioning

confidence: 99%

“…Previous sentiment studies took snapshots of attitudes towards COVID-19 at a particular point in time, making comparison of findings difficult. Some researchers did not clearly state their data collection period (e.g., Adamu et al, 2021). Thus far, our literature search did not find studies that took into consideration the timeline for the analysis of social media messages on COVID-19 vaccination.…”

Section: Introductionmentioning

confidence: 99%

“…The negative sentiments attracted more attention, but studies showed that there were more positive than negative views on vaccination. For example, Adamu et al (2021) reported that there were more neutral sentiments (47%) in tweets on COVID-19 vaccines than positive (31%) or negative (22%) sentiments. However, in January-February 2022, there were fewer neutral tweets (15.8%) and more positive (49.6%) than negative (34.6%) sentiments (Ong et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Have Your Say! Malaysian X (Twitter) Users Speak Their Minds About COVID-19 Vaccination

Ismail,

Ting

2024

Hum. Behav. Dev. Soc.

View full text Add to dashboard Cite

In this study, Malaysian X (formerly Twitter) users’ views were examined on COVID-19 vaccination. The specific objectives were to identify issues that were important to X users, and identify changes in views on COVID-19 vaccination. Tweets were collected from 1 January to 31 December 2021; altogether 5,766 tweets (199,900 words) were collected, and 150 tweets (5,200 words) were systematically selected for analysis. Thematic analysis showed that the tweets were more concerned about administration of the COVID-19 vaccine (56.7%) than its impact (35.3%) or COVID-19 control measures (8%). Positive sentiments increased during the 12 months. In Phase 1 (1 January–23 February 2021), the public were uncertain and sceptical while waiting for vaccination. In Phase 2 (24 February–24 September 2021) when vaccination was underway, the tweets reflected an informed stance, and X users were even proactive in promoting vaccination benefits and correcting misinformation. By Phase 3 (25 September–31 December 2021) when vaccination for teenagers and s booster shot program began, there was a dilemma of wanting to return to normal life vis-à-vis prioritizing health and safety. The study data indicated more anti- than pro-vaccination tweets, but the X community had self-correcting mechanisms when vaccine hesitancy surfaced.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Have Your Say! Malaysian X (Twitter) Users Speak Their Minds About COVID-19 Vaccination

Ismail,

Ting

2024

Hum. Behav. Dev. Soc.

View full text Add to dashboard Cite

show abstract

“…After the calculation of the distances, the K smallest distances will be obtained and it will verify how many classes are in the neighborhood. From the quantity measured by the algorithm, the class that appears the most is the one that should be assigned to the new data [22,23].…”

Section: Classification Algorithmsmentioning

confidence: 99%

Predictive Analysis of COVID-19 Symptoms in Social Networks through Machine Learning

2022

View full text Add to dashboard Cite

Social media is a great source of data for analyses, since they provide ways for people to share emotions, feelings, ideas, and even symptoms of diseases. By the end of 2019, a global pandemic alert was raised, relative to a virus that had a high contamination rate and could cause respiratory complications. To help identify those who may have the symptoms of this disease or to detect who is already infected, this paper analyzed the performance of eight machine learning algorithms (KNN, Naive Bayes, Decision Tree, Random Forest, SVM, simple Multilayer Perceptron, Convolutional Neural Networks and BERT) in the search and classification of tweets that mention self-report of COVID-19 symptoms. The dataset was labeled using a set of disease symptom keywords provided by the World Health Organization. The tests showed that Random Forest algorithm had the best results, closely followed by BERT and Convolution Neural Network, although traditional machine learning algorithms also have can also provide good results. This work could also aid in the selection of algorithms in the identification of diseases symptoms in social media content.

show abstract

“…From the publications (n = 47), eight studies compared machine learning models and chose the best-performing algorithm. In these comparisons, SVM [43,44,70] and LR [43,46,68] stood out with highly accurate performances. When comparing deep learning models and machine learning models, BERT showed an outstanding performance [51,55,64,69], whereas in comparison to deep learning models in [61], BiLSTM yielded the best performance.…”

mentioning

confidence: 92%

Role of Artificial Intelligence for Analysis of COVID-19 Vaccination-Related Tweets: Opportunities, Challenges, and Future Trends

et al. 2022

View full text Add to dashboard Cite

Pandemics and infectious diseases are overcome by vaccination, which serves as a preventative measure. Nevertheless, vaccines also raise public concerns; public apprehension and doubts challenge the acceptance of new vaccines. COVID-19 vaccines received a similarly hostile reaction from the public. In addition, misinformation from social media, contradictory comments from medical experts, and reports of worse reactions led to negative COVID-19 vaccine perceptions. Many researchers analyzed people’s varying sentiments regarding the COVID-19 vaccine using artificial intelligence (AI) approaches. This study is the first attempt to review the role of AI approaches in COVID-19 vaccination-related sentiment analysis. For this purpose, insights from publications are gathered that analyze the (a) approaches used to develop sentiment analysis tools, (b) major sources of data, (c) available data sources, and (d) the public perception of COVID-19 vaccine. Analysis suggests that public perception-related COVID-19 tweets are predominantly analyzed using TextBlob. Moreover, to a large extent, researchers have employed the Latent Dirichlet Allocation model for topic modeling of Twitter data. Another pertinent discovery made in our study is the variation in people’s sentiments regarding the COVID-19 vaccine across different regions. We anticipate that our systematic review will serve as an all-in-one source for the research community in determining the right technique and data source for their requirements. Our findings also provide insight into the research community to assist them in their future work in the current domain.

show abstract

Text Analytics on Twitter Text-based Public Sentiment for Covid-19 Vaccine: A Machine Learning Approach

Cited by 5 publications

References 18 publications

Have Your Say! Malaysian X (Twitter) Users Speak Their Minds About COVID-19 Vaccination

Have Your Say! Malaysian X (Twitter) Users Speak Their Minds About COVID-19 Vaccination

Predictive Analysis of COVID-19 Symptoms in Social Networks through Machine Learning

Role of Artificial Intelligence for Analysis of COVID-19 Vaccination-Related Tweets: Opportunities, Challenges, and Future Trends

Contact Info

Product

Resources

About