In this final chapter of the book, we will build an end-to-end streaming data pipeline that integrates Amazon ML within the Kinesis Firehose, AWS Lambda, and Redshift pipeline. We extend the Amazon ML capabilities by integrating it with these other AWS data services to implement real-time Tweet classification.
In a second part of the chapter, we show how to address problems beyond a simple regression and classification and use Amazon ML for Named Entity Recognition and content-based recommender systems.
The topics covered in this chapter are as follows:
- Training a twitter classification model
- Streaming data with Kinesis
- Storing with Redshift
- Using AWS Lambda for processing
- Named entity recognition and recommender systems
In the chapter's conclusion, we will summarize Amazon ML's strengths and weaknesses.