Skip to content

kaustubhgupta/Covid-Tweets-Analysis-Dashboard

Repository files navigation

Covid-Tweets-Analysis-Dashboard 📈📉📊

Covid-Tweets-Analysis-Dashboard

Live Notebook 😋

Binder

Related Article 🧾

Creating Interactive Jupyter Notebooks and Deployment on Heroku Using Voila (Towards Data Science)

Preview 📺

preview

Conclusions 😮

  • Mondays and Fridays were the most stressful days as there was less activity as compared to other days meaning that even if the work was done via homes, the working population was round the clock. This can be proved by the fact that the maximum of tweets was made on Saturdays!
  • March's last one-week tweets were greater than April's 10 days. This implies that when a country-wide lockdown was imposed on the 24th Midnight, there was panic among people and it is evident from the number of tweets made this week.
  • 11:00 am was the peak time for the users followed by 10:00 am and 12 pm.
  • #Corona was trending with #Coronavirus and #covid19. These are the obvious hashtags present in almost every tweet.
  • Most of the tweets were made from the top metropolitan cities such as New Delhi, Mumbai, Kolkata.
  • The most tagged person in tweets was the prime minister of India, Narendra Modi. People may want to show their concerns and problems directly with him. Arvind Kejriwal and Amit Shah also made it to the list
  • Most of the people on Twitter positively took the lockdown and around 44% of tweets were positive.
  • As the maximum number of tweets were made on Saturday, there was a slight peak in the number of negative tweets on Saturdays

Project Tech Stack 🏟

  • Python (Language)
  • Libraries
    • ipywidgets (Interactive elements)
    • textblob (Sentiment detection)
    • wordcloud (Wordcloud)
    • nltk (Text data processing)
    • numpy (Data Manipulation)
    • pandas (Data Manipulation)
    • matplotlib (Simple plotting)
    • plotly (Graphs)
    • voila (Notebook deployment)
  • binder (Notebook deployment server)