Streaming Twitter sentiment analysis
In this chapter, our main project consists of real-time sentiment classification of Tweets. This will allow us to demonstrate how to use an Amazon ML model that we've trained to process real-time data streams, by leveraging the AWS data ecosystem.
We will build an infrastructure of AWS services that includes the following:
- Amazon ML: to provide a real-time classification endpoint
- Kinesis firehose: To collect the Tweets
- AWS Lambda: To call an Amazon ML streaming endpoint
- Redshift: To store the Tweets and their sentiment
- S3: To act as a temporary store for the Tweets collected by Kinesis Firehose
- AWS Cloudwatch: To debug and monitor
We will also write the necessary Python scripts that feed the Tweets to Kinesis Firehose.
Popularity contest on twitter
All good data science projects start with a question. We wanted a social network question not tied to a current political or societal context. We will be looking into the popularity of vegetables on twitter. We want...
