A Review of Sentimental Analysis for Vaccine Dataset Using BI-LSTM Method DOI

Alok Soreng,

Safdar Sardar Khan, Arpit Deo

et al.

Published: Dec. 8, 2023

In the age of digital communication, realm public opinion, particularly on pressing issues like vaccination, has found its voice through myriad online platforms. Given surge discussions around vaccines, especially in wake COVID-19 pandemic, there is an imperative need to decipher underlying sentiments from these vast datasets. Sentiment analysis, a prominent branch Natural Language Processing (NLP), provides tools extract, process, and categorize such sentiments. While traditional machine learning models have held their ground sentiment analysis tasks, intricate nature human emotions language patterns demands more refined techniques. The Bidirectional Long Short-Term Memory (BI-LSTM) model, enhanced variant recurrent neural networks, emerges as promising candidate with capability capture contextual information both past future data points sequences. This review paper delves into application efficacy BI-LSTM method for vaccine-related Through comprehensive we evaluate performance metrics, benefits, limitations, positioning against other prevalent models. findings suggest that holds significant potential providing nuanced insights regarding which instrumental stakeholders craft informed strategies communications.

Language: Английский

Analysis of Pharmaceutical Companies’ Social Media Activity during the COVID-19 Pandemic and Its Impact on the Public DOI Creative Commons
Sotirios Gyftopoulos, George Drosatos, Giuseppe Fico

et al.

Behavioral Sciences, Journal Year: 2024, Volume and Issue: 14(2), P. 128 - 128

Published: Feb. 9, 2024

The COVID-19 pandemic, a period of great turmoil, was coupled with the emergence an "infodemic", state when public bombarded vast amounts unverified information from dubious sources that led to chaotic landscape. excessive flow messages citizens, combined justified fear and uncertainty imposed by unknown virus, cast shadow on credibility even well-intentioned affected emotional public. Several studies highlighted mental toll this environment took citizens analyzing their discourse online social networks (OSNs). In study, we focus activity prominent pharmaceutical companies Twitter, currently known as X, well public's response during pandemic. Communication between users is examined compared in two discrete channels, non-COVID-19 channel, based content posts circulated them March 2020 September 2022, while profile outlined through state-of-the-art emotion analysis model. Our findings indicate significantly increased channel predominant both channels joy. However, exhibited upward trend circulation quotes replies produced users, stark presence negative charge diffusion indicators, reveal preference for promoting tweets conveying charge, such fear, surprise, research study can inform development communication strategies emotion-aware future crises.

Language: Английский

Citations

4
Multimodal marvels of deep learning in medical diagnosis using image, speech, and text: A comprehensive review of COVID-19 detection DOI Creative Commons
Md. Shofiqul Islam, Khondokar Fida Hasan, Hasibul Hossain Shajeeb

et al.

AI Open, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

Citations

0
Clarifying Misunderstandings in COVID-19 Vaccine Sentiment and Stance Analysis and Their Implications for Vaccine Hesitancy: A Systematic Review DOI Creative Commons
Lorena Barberia, Belinda Lombard, Norton Trevisan Roman

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: March 25, 2025

Abstract Background Advances in machine learning (ML) models have increased the capability of researchers to detect vaccine hesitancy social media using Natural Language Processing (NLP). A considerable volume research has identified persistence COVID-19 discourse shared on various platforms. Methods Our objective this study was conduct a systematic review employing sentiment analysis or stance detection towards vaccines and vaccination spread Twitter (officially known as X since 2023). Following registration PROSPERO international registry reviews, we searched papers published from 1 January 2020 31 December 2023 that used supervised assess through Twitter. We categorized studies according taxonomy five dimensions: tweet sample selection approach, self-reported type, classification typology, annotation codebook definitions, interpretation results. analyzed if report different trends than those by examining how is measured, whether efforts were made avoid measurement bias. Results found bias widely prevalent analyze toward vaccination. The reporting errors are sufficiently serious they hinder generalisability these understanding individual opinions communicate reluctance vaccinate against SARS-CoV-2. Conclusion Improving NLP methods crucial addressing knowledge gaps discourse.

Language: Английский

Citations

0
Comparison of Impressions of COVID-19 Vaccinations Stratified by the Number of Vaccinations Among Japanese Healthcare Professional University Students DOI Open Access
Akihiro Yokoyama, Hiromi Suzuki, Hiroaki Kataoka

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: March 9, 2024

Objective: Coronavirus infectious disease, that emerged in 2019 (COVID-19) has been a major public health issue not only Japan, but also worldwide, and the implementation of proper vaccination strategy important. To promote vaccination, present study compared impressions COVID-19 vaccinations stratified by number among healthcare professional university students Okayama, suggests better strategies. Method: A total 212 Japanese were enrolled this clinical qualitative using text mining method. self-reported questionnaire, including questions such as "What do you think about vaccinations?" was performed. We examined vaccinations, sex, history infection, daily mask use. Results: 5,935 words obtained "Think" (169 times) most frequently used followed "Inject" (108 times), "Inoculation" (97 "Vaccine" (83 "Corona" (66 "Side effects" (49 times). Characteristic "Safety" non-vaccinated subjects "Necessary" vaccinated subjects. In addition, men "Frightening" women characteristic fundamental features. Conclusion: Impressions differed students. The provision appropriate information on safety to side effects appears be necessary. sex-specific may required for

Language: Английский

Citations

2
Comparing the accuracy of ANN with transformer models for sentiment analysis of tweets related to COVID-19 Pfizer vaccines DOI
Xuanyi Wu, Bingkun Wang, Wenling Li

et al.

Chaos Solitons & Fractals, Journal Year: 2024, Volume and Issue: 185, P. 115105 - 115105

Published: June 16, 2024

Language: Английский

Citations

2
COVID-19 Vaccine Hesitancy Among Healthcare Workers: A Phenomenological Study of Skepticism DOI Open Access

Parvathy Thampy,

Shweta Sharma,

Pragya Joshi

et al.

Cureus, Journal Year: 2024, Volume and Issue: unknown

Published: April 17, 2024

Rationale: Despite the prioritizing healthcare workers (HCWs) for COVID-19 in a systematized manner phenomenon of vaccine hesitancy was observed them. HCWs are presumed to be pre-emptive up-taking due their closest association and having reasonable background information. Hence, we intended explore investigate phenomenology skepticism toward among HCWs. Method: A sequential explanatory mixed methods study design incorporating baseline cross-sectional survey followed by qualitative semiquantitative text-mining approach adopted tertiary care center Madhya Pradesh, India. Six hundred seventy-nine quantitative data 30 interviews were surveyed. After determining quantum traits hesitant HCWs, participants purposively selected in-depth analysis based on grounded theory using framework consolidated from psychological philosophical plane skepticism. This complemented mono/bigram network plotting. Results: Approximately one-fifth (18%,122 out 679) initially, one-tenth initially (10 122) terminally hesitant. Hesitant non-hesitant similar except comorbidity status. Five themes emerged namely individual, vaccine-related, social, system, contextual after thematic consolidation. Words/phrases indicating individualistic desire knowing more, internal conflicts, conjecture mined further. The plot showed diversified expressions participants. Conclusion: There seems requirement prime offering objective information beforehand removing diffidence systematic addressing psychology prevalent partisan belief circumstances future.

Language: Английский

Citations

1
Challenges issues and future recommendations of deep learning techniques for SARS-CoV-2 detection utilising X-ray and CT images: a comprehensive review DOI Creative Commons
Md Shofiqul Islam, Fahmid Al Farid, F. M. Javed Mehedi Shamrat

et al.

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e2517 - e2517

Published: Dec. 24, 2024

The global spread of SARS-CoV-2 has prompted a crucial need for accurate medical diagnosis, particularly in the respiratory system. Current diagnostic methods heavily rely on imaging techniques like CT scans and X-rays, but identifying these images proves to be challenging time-consuming. In this context, artificial intelligence (AI) models, specifically deep learning (DL) networks, emerge as promising solution image analysis. This article provides meticulous comprehensive review imaging-based diagnosis using up May 2024. starts with an overview covering basic steps learning-based data sources, pre-processing methods, taxonomy techniques, findings, research gaps performance evaluation. We also focus addressing current privacy issues, limitations, challenges realm diagnosis. According taxonomy, each model is discussed, encompassing its core functionality critical assessment suitability detection. A comparative analysis included by summarizing all relevant studies provide overall visualization. Considering best deep-learning detection, conducts experiment twelve contemporary techniques. experimental result shows that MobileNetV3 outperforms other models accuracy 98.11%. Finally, elaborates explores potential future directions methodological recommendations advancement.

Language: Английский

Citations

1
Textual Emotion Analysis-based Disabled People Talking Using Improved Metaheuristics with Deep Learning Techniques for Intelligent Systems DOI Creative Commons
Haya Mesfer Alshahrani, Ishfaq Yaseen,

Suhanda Drar

et al.

Deleted Journal, Journal Year: 2023, Volume and Issue: 2(3)

Published: Sept. 15, 2023

Due to the complexity of generalizing and modeling series brain signals, detecting emotions in people with sensory disabilities still continues be challenging. Hence, brain–computer interface technology was used study behavior based on signals. Emotion analysis is a widely robust data mining method. It provides an excellent opportunity monitor, evaluate, determine, understand sentiments consumers respect product or service. Yet, recognition model visual has not been evaluated, even though previous studies have already proposed classification using machine learning approaches. Therefore, this introduces new salp swarm algorithm deep recurrent neural network-based textual emotion (SSADRNN-TEA) technique for disabled persons. The major intention SSADRNN-TEA focus detection that exist social media content. In work, undergoes preprocessing make input compatible latter stages processing BERT word embedding process applied. Moreover, network (DRNN) exploited. Finally, SSA exploited optimal adjustment DRNN hyperparameters. A widespread experiment involved simulating real-time performance experimental values revealed improved terms several evaluation metrics.

Language: Английский

Citations

3
Modified Aquila Optimizer with Stacked Deep Learning-Based Sentiment Analysis of COVID-19 Tweets DOI Open Access
Ahmed S. Almasoud,

Hala J. Alshahrani,

Abdulkhaleq Q. A. Hassan

et al.

Electronics, Journal Year: 2023, Volume and Issue: 12(19), P. 4125 - 4125

Published: Oct. 3, 2023

In recent times, global cities have been transforming from traditional to sustainable smart cities. text sentiment analysis (SA), many people face critical issues namely urban traffic management, living quality, information security, energy usage, safety, etc. Artificial intelligence (AI)-based applications play important roles in dealing with these crucial challenges SA. such scenarios, the classification of COVID-19-related tweets for SA includes using natural language processing (NLP) and machine learning methodologies classify tweet datasets based on their content. This assists disseminating relevant information, understanding public sentiment, promoting practices areas during this pandemic. article introduces a modified aquila optimizer stacked deep learning-based COVID-19 Classification (MAOSDL-TC) technique The presented MAOSDL-TC incorporates FastText, an effective powerful representation approach used generation word embeddings. Furthermore, utilizes attention-based bidirectional long short-term memory (ASBiLSTM) model sentiments that exist tweets. To improve detection results ASBiLSTM model, MAO algorithm is applied hyperparameter tuning process. validated benchmark dataset. experimental outcomes implied promising compared models terms different measures. improves accuracy interpretability prediction.

Language: Английский

Citations

3
Unveiling Sentiments in Javanese Text: A Study on Sentiment Analysis for the Javanese Language DOI

Raymasterio Vera Lucky,

Baladiffa Aurora Herkanyaka,

Vieri Ferdian Putra Basuki

et al.

Published: Oct. 18, 2023

The Javanese language faces challenges due to its complexity and decreasing usage as Indonesian becomes more prevalent. This study highlights the importance of preserving a crucial cultural heritage by developing sentiment analysis model tailored language. Two datasets, hotel reviews, "kuliah online" (Online Lectures) comments/tweets, were manually translated into three bilingual speakers, creating corpus. Through text preprocessing, word embedding, utilizing Bi-LSTM BERT methods. While Bi- LSTM + fastText achieved similar results in predicting negative positive sentiments. data comprising natural cases, mixed polarities, conflicting or target face for models. most successful model, BERT, an F1-score 77%. result obtained from will contribute preservation literature on NLP language, with future prospects involving training testing original datasets.

Language: Английский

Citations

0