You work in the data analytics department of a ride sharing software company. You need to use the K-means machine learning algorithm to separate your company’s optimized ride data into clusters based on ride coordinates. How would you use AWS Glue in the best way to build the data schema needed to classify the ride data?

Answer options:

A.Use Glue crawlers to crawl your ride share data.
B.Use Glue FindMatches to find and remove duplicate records in your data.
C.Use Glue to automatically generate code to classify the ride data based on coordinates.
D.Use Glue to transform and flatten your data so that you can classify the ride data based on coordinates.

Answer: A
Option Ais correct. The best way to build the schema for your data is to use a Glue crawler that leverages a classifier or multiple classifiers. (See the AWS Glue crawler https://docs.aws.amazon.com/glue/latest/dg/add-crawler.html documentation).
Answer B is incorrect because there is no stated need to remove duplicates from the data.
Option C is incorrect because you don’t need to automatically generate code since Glue will generate your schema for your data based on a prioritized list of classifiers without custom code (See the AWS Glue developers guide: (https://docs.aws.amazon.com/glue/latest/dg/add-classifier.html).
Option D is incorrect because there is no stated requirement to flatten the ride data.
Reference:
For an example, please see the AWS Machine Learning blog post titled Serverless unsupervised machine learning with AWS Glue and Amazon Athena: https://aws.amazon.com/blogs/machine-learning/serverless-unsupervised-machine-learning-with-aws-glue-and-amazon-athena/.