A Topic Modeling Comparison Between LDA, NMF, Top2Vec, and BERTopic to Demystify Twitter Posts

Frontiers in Sociology, Journal Year: 2022, Volume and Issue: 7

Published: May 6, 2022

The richness of social media data has opened a new avenue for science research to gain insights into human behaviors and experiences. In particular, emerging data-driven approaches relying on topic models provide entirely perspectives interpreting phenomena. However, the short, text-heavy, unstructured nature content often leads methodological challenges in both collection analysis. order bridge developing field computational empirical research, this study aims evaluate performance four modeling techniques; namely latent Dirichlet allocation (LDA), non-negative matrix factorization (NMF), Top2Vec, BERTopic. view interplay between relations digital media, takes Twitter posts as reference point assesses different algorithms concerning their strengths weaknesses context. Based certain details during analytical procedures quality issues, sheds light efficacy using BERTopic NMF analyze data.

Language: Английский

---

A performance comparison of supervised machine learning models for Covid-19 tweets sentiment analysis

PLoS ONE, Journal Year: 2021, Volume and Issue: 16(2), P. e0245909 - e0245909

Published: Feb. 25, 2021

The spread of Covid-19 has resulted in worldwide health concerns. Social media is increasingly used to share news and opinions about it. A realistic assessment the situation necessary utilize resources optimally appropriately. In this research, we perform tweets sentiment analysis using a supervised machine learning approach. Identification sentiments from would allow informed decisions for better handling current pandemic situation. dataset extracted Twitter IDs as provided by IEEE data port. Tweets are an in-house built crawler that uses Tweepy library. cleaned preprocessing techniques TextBlob contribution work performance evaluation various classifiers our proposed feature set. This set formed concatenating bag-of-words term frequency-inverse document frequency. classified positive, neutral, or negative. Performance evaluated on accuracy, precision, recall, F 1 score. For completeness, further investigation made Long Short-Term Memory (LSTM) architecture deep model. results show Extra Trees Classifiers outperform all other models achieving 0.93 accuracy score concatenated features LSTM achieves low compared classifiers. To demonstrate effectiveness set, with Vader technique based GloVe extraction

Language: Английский

---

Applications of artificial intelligence in battling against covid-19: A literature review

Chaos Solitons & Fractals, Journal Year: 2020, Volume and Issue: 142, P. 110338 - 110338

Published: Oct. 3, 2020

Language: Английский

---

Machine Learning in Healthcare Communication

Encyclopedia, Journal Year: 2021, Volume and Issue: 1(1), P. 220 - 239

Published: Feb. 14, 2021

Machine learning (ML) is a study of computer algorithms for automation through experience. ML subset artificial intelligence (AI) that develops systems, which are able to perform tasks generally having need human intelligence. While healthcare communication important in order tactfully translate and disseminate information support educate patients public, proven applicable with the ability complex dialogue management conversational flexibility. In this topical review, we will highlight how application ML/AI benefit humans. This includes chatbots COVID-19 health education, cancer therapy, medical imaging.

Language: Английский

---

The Longest Month: Analyzing COVID-19 Vaccination Opinions Dynamics From Tweets in the Month Following the First Vaccine Announcement

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 33203 - 33223

Published: Jan. 1, 2021

The coronavirus outbreak has brought unprecedented measures, which forced the authorities to make decisions related instauration of lockdowns in areas most hit by pandemic. Social media been an important support for people while passing through this difficult period. On November 9, 2020, when first vaccine with more than 90% effective rate announced, social reacted and worldwide have started express their feelings vaccination, was no longer a hypothesis but closer, each day, become reality. present paper aims analyze dynamics opinions regarding COVID-19 vaccination considering one-month period following announcement, until took place UK, civil society manifested higher interest process. Classical machine learning deep algorithms compared select best performing classifier. 2 349 659 tweets collected, analyzed, put connection events reported media. Based on analysis, it can be observed that

Language: Английский

---

Sentiment analysis on twitter tweets about COVID-19 vaccines usi ng NLP and supervised KNN classification algorithm

Indonesian Journal of Electrical Engineering and Computer Science, Journal Year: 2021, Volume and Issue: 23(1), P. 463 - 463

Published: July 1, 2021

The pandemic has taken the world by storm. Almost entire went into lockdown to save people from deadly COVID-19. Scientists around have come up with several vaccines for virus. Amongthem, Pfizer, Moderna, and AstraZeneca become quite famous. General however been expressing their feelings about safety effectiveness of on social media like Twitter. In this study, such tweets are being extracted Twitter using a API authentication token. raw stored processed NLP. data is then classified supervised KNN classification algorithm. algorithm classifies three classes, positive, negative, neutral. These classes refer sentiment general whose Tweets analysis. From analysis it seen that Pfizer shows 47.29%positive, 37.5% negative 15.21% neutral, Moderna 46.16%positive, 40.71% 13.13% 40.08%positive, 40.06% 13.86% neutral sentiment.

Language: Английский

---

Willingness of the Jordanian Population to Receive a COVID-19 Booster Dose: A Cross-Sectional Study

Vaccines, Journal Year: 2022, Volume and Issue: 10(3), P. 410 - 410

Published: March 9, 2022

SARS-CoV-2 (COVID-19) vaccines are critical for containing serious infections. However, as COVID-19 evolves toward more transmissible varieties and serum antibody levels in vaccinated persons steadily decline over time, the likelihood of breakthrough infections increases. This is a cross-sectional study based on an online questionnaire Jordanian adults (n = 915) to determine how individuals who have finished current vaccination regimen feel about prospective booster shot what factors might influence their decision. Almost half participants (44.6%) intended get dose vaccine. The most frequently mentioned reasons participants' reluctance vaccine were "The benefits not been scientifically proven" (39.8%), followed by "I took last short time ago, there will be no need take at least year" (24.6%). In turn, was infected with COVID-19; thus, I do require dose" reported reason (13.1%). These findings highlight considerable hesitancy immunization among Jordanians, well variables associated hesitancy, which aid creating excellent campaigns regarding doses.

Language: Английский

---

Arabic Tweets-Based Sentiment Analysis to Investigate the Impact of COVID-19 in KSA: A Deep Learning Approach

Big Data and Cognitive Computing, Journal Year: 2023, Volume and Issue: 7(1), P. 16 - 16

Published: Jan. 13, 2023

The World Health Organization (WHO) declared the outbreak of Coronavirus disease 2019 (COVID-19) a pandemic on 11 March 2020. evolution this has raised global health concerns, making people worry about how to protect themselves and their families. This greatly impacted people’s sentiments. There was dire need investigate large amount social data such as tweets others that emerged during post-pandemic era for assessment As result, study aims at Arabic tweet-based sentiment analysis considering COVID-19 in Saudi Arabia. datasets have been collected two different periods three major regions Arabia, which are: Riyadh, Dammam, Jeddah. Tweets were annotated with sentiments: positive, negative, neutral after due pre-processing. Convolutional neural networks (CNN) bi-directional long short memory (BiLSTM) deep learning algorithms applied classifying tweets. experiment showed performance CNN achieved 92.80% accuracy. BiLSTM scored 91.99% terms Moreover, an outcome study, overwhelming upsurge negative sentiments observed dataset compared before COVID-19. technique state-of-the-art techniques literature it proposed is promising various parameters.

Language: Английский

---

Deep Learning-Based Methods for Sentiment Analysis on Nepali COVID-19-Related Tweets

Computational Intelligence and Neuroscience, Journal Year: 2021, Volume and Issue: 2021, P. 1 - 11

Published: Jan. 1, 2021

COVID-19 has claimed several human lives to this date. People are dying not only because of physical infection the virus but also mental illness, which is linked people’s sentiments and psychologies. People’s written texts/posts scattered on web could help understand their psychology state they in during pandemic. In paper, we analyze sentiment based classification tweets collected from social media platform, Twitter, Nepal. For this, we, first, propose use three different feature extraction methods—fastText-based (ft), domain-specific (ds), domain-agnostic (da)—for representation tweets. Among these methods, two methods (“ds” “da”) novel used study. Second, convolution neural networks (CNNs) implement proposed features. Last, ensemble such CNNs models using CNN, works an end-to-end manner, achieve end results. evaluation CNN models, prepare a Nepali Twitter dataset, called NepCOV19Tweets, with 3 classes (positive, neutral, negative). The experimental results dataset show that our possess discriminating characteristics for classification. Moreover, impart robust stable performance Also, can be as benchmark study COVID-19-related analysis language.

Language: Английский

---

TClustVID: A novel machine learning classification model to investigate topics and sentiment in COVID-19 tweets

Knowledge-Based Systems, Journal Year: 2021, Volume and Issue: 226, P. 107126 - 107126

Published: May 6, 2021

Language: Английский

---