Pulling Twitter Data Using Python

Introduction

• Welcome to the tutorial on pulling Twitter data using Python.
• Useful for practicing data analysis and machine learning.

Prerequisites

• Ensure Python is installed.
  • Check a tutorial for installation if not yet done.

Getting Twitter API Access

1. Obtain a Twitter Token:

   • Go to apps.twitter.com
   • Log in with your Twitter account; sign up if you don’t have one.

2. Create an App:

   • Click on "create app."
   • Fill in the application details:
     • Name: Your choice
     • Description: Your choice
     • Website URL: Use placeholder if no website.
   • Describe your intention with the Twitter API (e.g., for study purposes).
   • Press the create button.

3. Application Approval:

   • Twitter will assess your application (approval typically within a day).
   • Check your email for approval notification.

4. Get API Keys and Tokens:

   • Click on the app details.
   • Navigate to "keys and tokens" section.
   • Note down the API key, API secret key, access token, access token secret.

Setting Up in Jupyter Notebook

1. Import Tweepy Library:

   import tweepy
   

   • If an error occurs, install the library using:

     pip install tweepy
     

2. Store API Credentials:

   • Set up the following variables:

     api_key = 'your_api_key'
     api_secret_key = 'your_api_secret_key'
     access_token = 'your_access_token'
     access_token_secret = 'your_access_token_secret'
     

3. Authenticate with Twitter API:

   auth = tweepy.OAuthHandler(api_key, api_secret_key)
   auth.set_access_token(access_token, access_token_secret)
   api = tweepy.API(auth)
   

Pulling Data

Based on User Timeline

• To get tweets from a user:

  userResults = api.user_timeline(id='username', count=10)
  

• Print the tweets:

  for tweet in userResults:
      print(tweet.text)
  

Based on Search Query

• To search for tweets containing a specific keyword:

  hasilSearch = api.search(q='jakarta', lang='id', count=10)
  

• Print the search results:

  for tweet in hasilSearch:
      print(f'{tweet.user.screen_name} tweeted: {tweet.text}')
  

Conclusion

• Successfully pulled data based on user and search queries.
• Next video will cover processing the pulled data for advanced output.
• Viewers are encouraged to like, subscribe, and ask questions in the comments.