Dataset

This dataset was created by Mahendra Ch

Contents

Context

I collect recent tweets about the COVID-19 vaccines used in entire world on large scale, as following: * Pfizer/BioNTech;
* Sinopharm;
* Sinovac;
* Moderna;
* Oxford/AstraZeneca;
* Covaxin;
* Sputnik V.

Data collection

The data is collected using tweepy Python package to access Twitter API. For each of the vaccine I use relevant search term (most frequently used in Twitter to refer to the respective vaccine)

Data collection frequency

Initial data was merged from tweets about Pfizer/BioNTech vaccine. I added then tweets from Sinopharm, Sinovac (both Chinese-produced vaccines), Moderna, Oxford/Astra-Zeneca, Covaxin and Sputnik V vaccines. The collection was in the first days twice a day, until I identified approximatively the new tweets quota and then collection (for all vaccines) stabilized at once a day, during morning hours (GMT).

Inspiration

You can perform multiple operations on the vaccines tweets. Here are few possible suggestions:

• Study the subjects of recent tweets about the vaccine made by various producers;
• Perform various NLP tasks on this data source (topic modelling, sentiment analysis);
• Using the COVID-19 World Vaccination Progress (where we can see the progress of the vaccinations and the countries where the vaccines are administered), you can study the relationship between the vaccination progress and the discussions in social media (from the tweets) about the vaccines.

This dataset contains Twitter posts containing daily updates of location-based COVID–19 vaccine-related tweets from January 2021 to August 2021.

With an existing Twitter account, we applied for Developer Access and were granted access to Twitter Academic Researcher API which allows for over 10 million tweets per month. Then, we created an application to generate the API credentials (access tokens) from Twitter. The access token was used in Python (v3.6) script to authenticate and establish a connection to the Twitter database. To get goe-tagged vaccine-related tweets, we used the python script we developed to perform a historical search (archive search) of vaccine-related keywords with place country South Africa (ZA). By goe-tagged tweets, we refer to Twitter posts with a know location. These vaccine-related keywords include but are not limited to the vaccine, anti-vaxxer, vaccination, AstraZeneca, Oxford-AstraZeneca, IChooseVaccination, VaccineToSaveSouthAfrica, JohnsonJohnson, and Pfizer. The keywords were selected from the trending topic during the period of discussion. A complete list of the keywords is shown below:

Oxford-AstraZeneca, AstraZeneca, JohnsonJohnson, Vaccine, BioNTech, anti-vaccine, jab, Vaccination, Covax, Vaccine Rollout, Sputnik, VaccineToSaveSouthAfrica, IChooseVaccination, TeachersVaccine, AstraZeneca vaccine, Pfizer, J & J, Johonson & Johnson, Moderna, VaccinesWork, VacciNation, Vaccine, Steriod, COVIDvaccine, covax, VaccineEquity, VaccineReady, Jab OR PfizerGang, Scamdemic, Plandemic, Scaredemic, COVID-19, coronavirus, SARS-CoV-2, anti-vaxxers, jab, Pfizer, BioNTech, JJ, Vaccine, JohnsonJohnson Vaccine, Vaccine Rollout, J & J, Sputnik, COVAX, CoronaVac

The preferred language of the tweet is English.

Twitter Pfizer - AntiCovid

Dataset Description

This dataset contains tweets collected from Twitter concerning the Pfizer COVID-19 vaccine. The primary goal of this dataset is to help researchers, data scientists, and analysts understand public sentiment regarding the Pfizer vaccine. It is structured to support sentiment analysis, subjectivity analysis, and classification tasks.

Column Descriptions

1. id:
2. Description: A unique identifier for each tweet.
   • Type: Integer.

3. Purpose: Used to uniquely identify and differentiate each tweet in the dataset.

4. Text:

   • Description: The content of the tweet in textual form.
   • Type: String.
   • Purpose: Provides the actual content of each tweet. This column can be used for various natural language processing (NLP) tasks such as sentiment analysis, keyword extraction, and topic modeling.

5. Subjectivity:

   • Description: A score between 0 and 1 that represents the degree of subjectivity in the tweet. A score of 0 means the tweet is objective, while a score of 1 indicates it is highly subjective.
   • Type: Float (range: 0 to 1).
   • Purpose: Indicates how much of the tweet is based on personal opinion versus factual information. This can be useful for identifying tweets that are more opinion-driven.

6. Polarity:

   • Description: A sentiment polarity score, ranging from -1 (extremely negative) to 1 (extremely positive).
   • Type: Float (range: -1 to 1).
   • Purpose: Used to determine the emotional tone of the tweet, whether it is negative, neutral, or positive. This column is key for sentiment analysis tasks.

7. Target:

   • Description: A binary variable indicating the overall sentiment of the tweet towards the Pfizer vaccine. A value of 0 indicates a negative sentiment, and a value of 1 indicates a positive sentiment.
   • Type: Integer (0 or 1).
   • Purpose: Serves as the target label for machine learning models. It helps to classify the sentiment of the tweets as positive or negative.

Provenance

This dataset was collected from Twitter using the Twitter API. Tweets containing keywords and hashtags related to the Pfizer vaccine and COVID-19 were gathered over a specified time frame. The subjectivity and polarity scores were calculated using the TextBlob library in Python, which is widely used for natural language processing tasks.

The data has been organized to ensure compatibility with a wide range of NLP and machine learning tasks. Please note that this dataset is anonymized, and any sensitive personal information has been removed to adhere to data privacy standards.

Update Frequency

This dataset is static and will not be updated regularly. However, future versions may be released based on new tweet collections or to improve the quality of the dataset. If there are any significant updates, they will be indicated in the dataset’s metadata tab on Kaggle.

Context

Collected recent tweets about the COVID-19 vaccines used in the entire world on large scale, as follows:

• Pfizer/BioNTech;
• Sinopharm;
• Sinovac;
• Moderna;
• Oxford/AstraZeneca;
• Covaxin;
• Sputnik V.

Content

Starting with the step of loading the data using pandas, some basic data frame operations allow us to see that, for each tweet, all of the following information is available:

information about the user who tweeted

• user_name: Twitter handle
• user_location: where in the world the person tweets from (NOTE: there is no validation here… “your bed” is technically acceptable)
• user_description: user-written biography
• user_created: when they created their Twitter account
• user_followers: number of followers
• user_friends: number of accounts the user is following
• user_favourites: number of tweets the user has liked
• user_verified: indicates if the user is a well-known figure (boolean)

information about the tweet itself

• id: indexing value for Twitter API
• date: a DateTime object in the form of YYYY-MM-DD HH:MM:SS
• text: the tweet itself (MOST IMPORTANT)
• hashtags: list of hashtags used in the tweet (without ‘#’ character)
• source: which device was used for the tweet
• retweets: number of retweets received at the time the data was collected
• favorites: number of likes received at the time the data was collected
• is_retweet: indicates if the tweet is original or a retweet (boolean)

Data collection

The data is collected using tweepy Python package to access Twitter API. For each of the vaccine I use a relevant search term (most frequently used in Twitter to refer to the respective vaccine).

If you find this dataset useful, please consider upvoting ❤️

**Dates: **Nov 15 to Dec 16, 2021 Total Records: estimated 3,000,000 **Search Query: **vaccine OR vaccinemandate OR "vaccine mandate" OR pfizer OR moderna OR mRNA Data source: Twitter public API **Notes: **Data downloaded in CSV format and unedited for research use. Collection method: Netlytic

Please cite data: Li, E Rosalie. Dec 2021. Vaccine tweets 1-30. Hoaxlines disinformation database from Novel Science. https://www.kaggle.com/hoaxlines/vaccine-tweets

this is the dataset which contain four attribute Username Description Text Hashtags for Pfizer vaccination

Context

Winter Storm Uri in February 2021 caused havoc across the United States and specifically to Texas involving mass power outages, water and food shortages, and dangerous weather conditions.

This dataset consists of 23K+ tweets during the crisis week. Data is filtered to mostly include the tweets from influencers (users having more than 5000 followers) however there is a small subset of tweets from other users as well.

My notebook - https://www.kaggle.com/rajsengo/eda-texas-winterstrom-2021-tweets

Acknowledgements

• https://www.kaggle.com/gpreda/pfizer-vaccine-tweets - For the inspiration
• https://github.com/dataquestio/twitter-scrape - Reference utility to scrape twitter

Inspiration

Apply NLP techniques to undestand user sentiments about the crisis management