Each subdirectory contains sample code for using Amazon Machine Learning. Refer to the README.md file in each sub-directory for details on using each sample.

These samples show how to use the Amazon Machine Learning API for a targeted marketing application. This follows the "banking" dataset example described in the Developer Guide. There are three versions available:

• Targeted Marketing with Machine Learning in Java
• Targeted Marketing with Machine Learning in Python
• Targeted Marketing with Machine Learning in Scala

This sample application shows how to use Amazon Mechanical Turk to create a labeled dataset from raw tweets, and then build a machine learning model using the Amazon Machine Learning API that predicts whether or not new tweets should be acted upon by customer service. The sample shows how to set up an automated filter using AWS Lambda that monitors tweets on an Amazon Kinesis stream and sends notifications whenever the ML Model predicts that a new tweet is actionable. Notifications go to Amazon SNS, allowing delivery to email, SMS text messages, or other software services.

• Machine-Learning based Social Media Filtering (Python & JavaScript)

These samples show how to use the Amazon Machine Learning API to make real-time predictions from a mobile device. There are two versions available:

• Real-time Machine Learning Predictions from iOS
• Real-time Machine Learning Predictions from Android

This sample shows how to use the Amazon Machine Learning API to evaluate ML models using k-fold cross-validation.

• K-fold Cross-validation Sample (Python)

A collection of simple scripts to help with common tasks.

• Machine Learning Tools (python)

For assistance with using the Amazon Machine Learning Service, or these samples, please see the AWS Forums.