Scripts to find the most commonly followed accounts by a group of accounts on Twitter
Usage
First find a Twitter list for accounts you want to analyze. For example, the New Yorks Times maintains a list of NYT journalists on their official Twitter account. For this example, the username is nytimes and the list name is nyt-journalists.
Next, in the .env file, replace the sample variables from .env.example with the ones you want to use.
To get Twitter keys, visit https://apps.twitter.com/.
The results are saved to an S3 bucket, so that this container can be run anywhere (for example, as a task on ECS). You may need to log in to your AWS account and setup a user and IAM policy that has write access to the S3 bucket.
TWITTER_CONSUMER_KEY=xxx # Grab these from the Twitter developer console
TWITTER_CONSUMER_SECRET=xxx
TWITTER_ACCESS_TOKEN=xxx
TWITTER_ACCESS_TOKEN_SECRET=xxx
RESULTS_S3_BUCKET=xxx # Name of the S3 bucket where results should be uploaded
RESULTS_S3_BUCKET_PATH=results/ # Add the path in the bucket where the results should go
AWS_ACCESS_KEY_ID=xxx # AWS access keys that have permission to upload to the bucket
AWS_SECRET_ACCESS_KEY=xxxBuild the container with docker build -t twitter-most-followed-scripts . or use the container that is already built at tylerpearson/twitter-most-followed-scripts:latest.
Pass the username and list name as args to run the docker container:
docker run -d --env-file=.env \
tylerpearson/twitter-most-followed-scripts:latest \
username listnameBlog Post
A write up of the results found during an analysis of U.S. and Canadian Olympians can be viewed here.
Results!
A Jekyll template that can be used to displayed the results is available in the twitter-most-followed-site repo.
Here are some results: shown through Hifi templates.
License
MIT.