Twitter Treasures

This repository contains the code base to extract tweets from Twitter using Twitter API , analyse and visualise tweets for discovering information hidden in plain sight (data , opinion mining and sentiment analysis).

How to Start

Functions

1. Login funcitonalities
   • Username and Password
2. Sign Up functionalities

• Username and Password
• Twitter API : Consumer Key and Secret Key

3. View top trending tweets (Worldwide and specific countries)

• Provides Top 50 worldwide trending tweets
• Dataframe with Topic, Tweet Volume and URL
• Trending topics Wordcloud

4. Extract Tweets with Recent search Twitter API

• From the current day to last 7 days
• Advance Twitter Search Query can be used

5. Analysis of Twitter data

• Tweets preprocessing
• Metadata analysis (Wordclouds and Bar plots)
  • Source
  • Location
  • Language
• Tweets and User description
• Ngram Analysis (Line and Bar plots) -Unigram
  • Bigram
  • Trigram

• Sentiment Analysis (Naive bayes Algorithm) - Basic Approach
  

  • Hashtag Analysis (Bar plots and Wordcloud)
  • Username Analysis (Bar plots and Wordcloud)
  • Twitter sentiments visualisations (Bar plots , Wordcloud , Interactive HTML graphs for sentiments visual)
  • Sentiments topic analysis(Interactive HTML graphs for sentiments visual)

7. Disqus form for better feedbacks and resolving errors.

• Mail address is also provided for better accessibility

Rapid Prototyping Pipeline

The amount of time it took for me to build the first version of the usable application was just 2 days from conceptualisation of the idea to deployment. This was my first time development of a usable application.Streamlit is a must know tool for all python based Data App developers.

Heroku deployment

To Do

2. Add description on how to interpret the graphs
3. Structure the analysis based on a narration (!)
4. Named Entity Analysis
5. Correlation (?) (But how?)