Chapter 5. Explore the Dataset

In the previous chapter, we demonstrated how to ingest data into the cloud with Amazon Athena and Redshift. Amazon Athena offers ad hoc, serverless SQL queries for data in S3 without needing to set up, scale, and manage any clusters. Amazon Redshift provides the fastest query performance for enterprise reporting and business intelligence workloads—particularly those involving complex SQL with multiple joins and subqueries across many data sources, including relational databases and flat files. We created a data-catalog mapping for our S3-based data lake in S3 using AWS Glue Catalog. We ran ad hoc queries on our data lake with Athena. And we ran queries on our data warehouse with Amazon Redshift.

We also had a first peek into our dataset. As we’ve learned, the Amazon Customer Reviews Dataset consists of more than 150+ million of those customer reviews of products across 43 different product categories on the Amazon.com website from 1995 until 2015. The dataset contains the actual customer reviews text together with additional metadata. It comes in two formats: row-based tab-separated values (TSV) and column-based Apache Parquet.

In this chapter, we will use the SageMaker Studio integrated development environment (IDE) as our main workspace for data analysis and the model development life cycle. SageMaker Studio provides fully managed Jupyter Notebook servers. With just a couple of clicks, we can provision the SageMaker Studio IDE and start using Jupyter ...

Get Data Science on AWS now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.