SENTIMENTAL ANALYSIS OF COVID-19 TWITTER DATA USING DEEP LEARNING AND MACHINE LEARNING MODELS

The novel coronavirus disease (COVID-19) is an ongoing pandemic with large global attention. However, spreading fake news on social media sites like Twitter is creating unnecessary anxiety and panic among people towards this disease. In this paper, we applied machine learning (ML) techniques to predict the sentiment of the people using social media such as Twitter during the COVID-19 peak in April 2021. The data contains tweets collected on the dates between 16 April 2021 and 26 April 2021 where the text of the tweets has been labelled by training the models with an already labelled dataset of corona virus tweets as positive, negative, and neutral. Sentiment analysis was conducted by a deep learning model known as Bidirectional Encoder Representations from Transformers (BERT) and various ML models for text analysis and performance which were then compared among each other. ML models used were Naïve Bayes, Logistic Regression, Random Forest, Support Vector Machines, Stochastic Gradient Descent and Extreme Gradient Boosting. Accuracy for every sentiment was separately calculated. The classification accuracies of all the ML models produced were 66.4%, 77.7%, 74.5%, 74.7%, 78.6%, and 75.5%, respectively and BERT model produced 84.2 %. Each sentiment-classified model has accuracy around or above 75%, which is a quite significant value in text mining algorithms. We could infer that most people tweeting are taking positive and neutral approaches. © 2023, Universidad Politecnica Salesiana. All rights reserved.

Sentiment Analysis of COVID-19 Vaccines

Social media is invariably being used these days for exchanging information and views on global affairs including COVID-19 pandemic. In this study, we have worked to understand the public attitudes of people in different countries towards COVID-19 vaccines using social media platform Twitter. We have applied natural language processing techniques of sentiment analysis to get an insightful outlook on people's views. Hence, we categorized our results into fine-grained polarities to grasp the exact sentiment. For analyzing the sentiments, we have taken tweets that expressed sentiments for all countries, as well as for four countries that had higher fatality rates are United States of America, Mexico, Brazil and India. The people have expressed a neutral opinion towards the vaccines. Based on the sentiment, the vaccines were also ranked in which the people have expressed more faith in Sputnik V and Covishield vaccines. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Opinion classification at subtopic level from COVID vaccination-related tweets.

Coronavirus disease 2019 (Covid-19) is a contiguous disease which affected a large volume of population with a high mortality rate across the globe. For dealing with the recent spread of COVID-19, one of the prime measures was to vaccinate people in full extent. People across the globe have diverse opinion regarding the vaccination process, its side effect and effectiveness. Such opinions get located into different micro-blogging sites including twitter. Opinion mining through analyzing public sentiments of such micro-blogs is a common method for detection of public responses. This paper focuses on classifying the public opinions expressed related to COVID-19 vaccination at sub topic level. The procedure tries to find out different keywords regarding positive, negative and neutral sentences. From those keywords, different related query set was constructed using Rocchio query expansion algorithm for positive, negative and neutral sentiments. Later Extended query set is used to form subtopic using LDA algorithm to identify the nature of the tweets. The proposed LDA model came across with 0.56 coherence score with twenty subtopics, which is fair enough to classify the tweets in different classes. This trained model is finally used to classify the tweets in real time with Apache Kafka framework regarding different subtopic based on positive, negative or neutral sentiment.

Textual Dissection of Live Twitter Reviews on Corona Vaccines using Various Machine Learning Algorithms

The most popular hash tag on Twitter in 2020 was #COVID19 Vaccination, which got roughly 400 million notices. In this paper, we examine a worldview for unearth the feeling about COVID-19 inoculations among the public from Twitter. After obtaining the misconceptions and ideas in circulation, we suggest a solution for the same through Machine Learning algorithms. Twitter is a well known microblogging social media website where users distribute their perspectives on any topic(s). The ideology of textual dissection describes how people think about a text. It's the process of categorising tweets into positive and negative groups. Tweepy and TextBlob are Python libraries that can be used to extract and classify Tweets using Machine Learning methods including Naive Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Decision Tree. The goal is to make analysis, summarization, and classification as straightforward as possible. These computations comprise a positive, negative, or neutral assessment of Twitter data. In light of public perception, we hypothesize the best immunization feasible with maximum antibodies based on public perception through opinion research. © 2022 IEEE.

Twitter Sentiment Analysis on Approval of Covaxin for Children in India

Covid-19 and its different variants are still a big issue the whole world is facing right now. At present different SARS-CoV-2 vaccines are playing vital role to combat the coronavirus. The objective of this paper is to perform sentiment analysis on approval of Bharat Biotech covaxin for emergency use for children. The presented paper emphasizes on the sentiment analysis of tweets of the microblogging site Twitter. Python programming language with Natural Language processing toolkit (NLTK), TextBlob library and tweepy twitter API are used for the process. Machine learning algorithms are used for the classification of tweeets. Graphical representation has been used for the representation of the data after sentiment analysis based on hashtags. © 2022 IEEE.

Long Covid Sentiment Analysis of Twitter Posts to understand public concerns

Even as more people are getting vaccinated and measured steps with caution are taken to return towards normalcy, Long Covid still persists. Long Covid is a post-Covid condition in which patients still have symptoms for weeks or months after they have recovered from Covid-19. The Covid-19 epidemic and its accompanying societal mitigation methods, such as lockdowns, have resulted in a spike in people's usage of social media platforms like Twitter for conveying their views, opinions, and anxieties. As a result, in this paper, I have performed sentimental analysis on three sets of data that were collected relating to Long Covid, Long Covid in Kids and lastly Treating Long Covid. This was done to explore societal-scale reactions for the illness Long Covid. A total of 98386 tweets were extracted for the period of 11th December 2021 to 20th December 2021 using python's Tweepy package. After performing all the pre-processing on the tweets, a total of 15827 tweets were analyzed. AFFIN lexicon model was employed for performing sentimental analysis on the user's tweets. Visualizations in the form of bar charts, histograms, strip plots, box plots, pie charts, and word clouds have been created for gaining deeper insight into the sentiments of the tweets posted. The results showed that 44% of tweets are negative, 34 % of tweets are positive and 23 % of tweets are neutral for Long Covid. 39% of tweets are negative, 33 % of tweets are positive and 28% of tweets are neutral for the data set of Long Covid in Kids. These results show that negative sentiments outweigh positive sentiments relating to Long Covid. However, 41 % of tweets are positive, 32% of tweets are neutral and 27% of tweets are negative for the data set of treating Long Covid. This result portrays that people have more positive sentiments regarding treating Long Covid. © 2022 IEEE.

Opinion Mining Approach: To Understand Public Reactions towards Reopening of Schools and Colleges/Universities in India

The psychological, developmental, and educational consequences of the prolonged shutdown and digital shift to e-learning methods are being felt in every family throughout India. Amid the COVID-19 pandemic, the use of social media platforms such as Twitter has increased as people turn to it to express their thoughts, anxieties, and concerns. As a result, I conducted sentiment analysis on two sets of data related to the reopening of schools in India and the reopening of colleges/universities in India. The tweets were retrieved from the social media site Twitter using 9 keywords relating to the reopening of schools and colleges/universities. The Tweepy module in Python was used for the extraction process. A total of 2,494 tweets over the course of 16 days, from December 8th, 2021 to December 23 rd, 2021 were gathered. Later a total of 1,510 tweets were evaluated once all of the pre-processing steps were done. The sentimental analysis was carried out using 2 unsupervised Lexicon-based models-the Vader Lexicon and the SentiWordNet Lexicon method. In this study, the outcomes of both techniques are highlighted. The main aim of this research is to gauge the sentiments of the public reactions expressed towards the reopening of schools and colleges/universities. These findings may be a beneficial aspect to consider while the policy-making process. The results of this study suggest that by using both approaches there was a higher number of positive tweets which may indicate that citizens in India are in the favor of schools and colleges reopening. This study can further be expanded by incorporating public viewpoints expressed on various other social networking platforms and can be extended to create a comprehensive dashboard with state-wise catered guidance, instructions, and resources for safely reopening schools and colleges in India. © 2022 IEEE.

Sentiment Analysis of Corona Vaccination on Twitter Data using Machine Learning Techniques

Currently, most of the people are stating their opinion through social media. Public opinion stands an important one while reviewing any product, Movie etc. The field of opinion mining has turn out to be the most significant areas of Natural Language Processing. Opinion mining can be used to make decisions about the product or Movie centred on the reviews given by customers in social media. Twitter is one of the most popular social media platforms;Twitter allows users to express their opinions through tweets. It is vital to use these reviews to make decisions and manage the situation. This paper examines public opinions regarding Corona vaccination during pandemic situations with the goal of exploring the opinions of people. In this work, opinion mining of tweets posted by users in Twitter regarding corona vaccination has been performed using Machine Learning Models and Lexicon based approach. Data are collected from Twitter through Tweepy API, and it is pre-processed using NLTK library using python. A comparison of eleven different classification algorithms is made to determine which the best is. As a result, we can conclude that people support neutral decision to take Corona vaccine during the current pandemic. © 2022 IEEE.

Supervised Multimodal Emotion Analysis of Violence on Doctors Tweets

With the advent of novel coronavirus pandemic doctors, health workers, and the government too, are trying their best of their capacity to deal with contemporary situations. It is genuine that when a person's close one is lost, they will react vociferously but accusing the doctors and workers and harming them is also morally indignant as the person saving so many lives his/her own life is in danger. With the boom of technology and how the world has come so close on social media, many social media users are expressing their views in either the support or opposition of the saviors of this pandemic, the doctors and the health care workers. These views of people are enough to create a good or bad impression of any doctor in minds of people and can even create a hostile behavior for that doctor by others, analyzing the stand of the person towards the ongoing violent situation towards workers using a multimodal emotional analysis combining both visual and textual data. This paper uses a Multimodal Transformer model which combines both visual and textual data is the sole purpose of this paper. Apart from the main aim, the paper will also explain whether in social media more information has been carried out by a text or more information can spread through images posted on social media. The paper will explain the use of appropriate loss function for imbalanced data also.

Sentiment Classification and Comparison of Covid-19 Tweets During the First Wave and the Second Wave Using NLP Techniques and Libraries

This research focuses on analysing the sentiments of people pertaining to severe periodic outbreaks of COVID-19 on two junctures – First Wave (Mar’20 & Apr’20) and Second Wave (Jun’21 & Jul’21)-since the first lockdown was undertaken with a view to curb the vicious spread of the lethal SARS-Cov-2 strain. Primarily, the objective is to analyse the public sentiment – as evident in the posted tweets - relating to the different phases of the pandemic, and to illuminate how keeping an eye on change in the tenor and tone of discussions can help government authorities and healthcare industry in raising awareness, reducing panic amongst citizens, and planning strategies to tackle the monumental crisis. Considering the daily volume of social media activity, in our project, we scoped to analyse the Tweets related to the two different pandemic stages – The First wave and the Second wave – by implementing Text Mining and Sentiment Analysis, subfields of Natural Language Processing. To manually extract tweets from the platform, we used Twitter API coupled with Python’s open-source package using a set of COVID-19-related keywords. Crucially, before finalising the project pipeline, we conducted a thorough secondary research to find the solutions and methodologies implemented in our area of interest. We listed the current works and attempted to plug the gaps in those via our experiment. We used several classification and boosting algorithms to create a framework to distinguish the textual data of the tweets. Relevant scope, limitations, and room for improvements have been discussed comprehensively in the upcoming sections. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Sentiment analysis on covid-19 vaccine using Twitter data: A NLP approach

Sentiment analysis is the process of mining the perception of people towards a service, product, policy or imminent issue from textual data. In this project, tweets relevant to Covid-19 Vaccine are extracted utilizing the Tweepy library. Next, tweet texts are converted into usable form in order to do sentiment analysis. After this, SentiWordNet lexicon is used to label the sentiment of the tweets. Stop words removal, Lemmatizing, stemming operations are also applied on the COVID-19 Vaccine tweets text data. Count Vectorizer and Tfidf Vectorizer are applied for mathematical conversion of the preprocessed text. Then, nine classification techniques namely - Multinomial-NB, Bernoulli-NB, Logistic-Regression, Ridge Classifier, Passive-Aggressive-Classifier, Perceptron, Random Forest classifier, AdaBoostClassifier and Linear SVM are applied on the dataset obtained for sentiment classification and results are obtained in terms of accuracy. The best cross validation test score obtained is 0.785 with Logistic Regression Classifier and TfidfVectorizer. © 2021 IEEE.