Indian Elections, Text-Mining Group-Project

Text Mining Data from Twitter APIs

Goal

Create a scripit using Python that extract Tweets related to the prime minister of India, "Narendra Modi" and get Insights about his popularity.

Plan

• Extract Tweets of #Modi and Modi account from last 14 days using Python and Twitter APIs.
• Create a Data Frame using Panda Library to store the Tweets.
• Clean the data removing uncessary characters.
• Munging the data and slice it into the columns (user, tweet, date, likes, retweets and len)
• Classifying the sentiment polarity of tweets using TextBlob Library.
• Peform a Sentiment Analysis using TextBlob and Panda Library.
• Use WorldCloud library To show most frequently words.
• Perform Visualization methods (pie Chart, scatter plot, bar plot)

Findings

Based on Modi’s Popularity on Twitter, he is front and center in upcoming election

• Only 8% of his tweets has negative sentiment, doing fairly good in writing tweets.
• 32% of twitter users have positive sentiment for him.
• Roughly 20% of users are against him. This may include the opposition.

Comparison of our Polarity PieChart against the results of recent Election in India

Visualizations