Google is hoping to entice enterprises to use its cloud platform and machine learning technologies to build online recommendation engines for their websites.

To do so, it’s just posted the following tutorial for a house-renting website that offers step-by-step instructions for developers looking to build an engine that can recommend homes to users based on their previous behavior.

Matthieu Mayran, a cloud solutions architect at Google, said he hoped to show developers how easy it is to use open-source technologies to create a simple recommendation engine on Google’s cloud.

“We hope that this solution will give you the nuts and bolts you need to build an intelligent and ever-improving application that makes the most of the information that your users give you,” Mayran wrote in his blog.

In the tutorial, Mayran’s sample system relies on a front-end that captures user interaction data and a permanent storage system to collect that data. The machine learning component of the system is based on Google’s recently released Cloud Dataproc service for managing data sets on Hadoop and Spark, and a second front-end storage system that’s designed for use in real-time.

Google’s tutorial also provides some tips on the considerations that developers need to keep in mind, like filtering methods and timeliness concerns.

Google also posted a second tutorial designed to show developers how they can harness TensorFlow and the Google Cloud Platform to carry out “fast, interactive data analysis and machine learning using Big Data sets”. For this one, Google has provided about six year’s of financial time series data from eight stock markets for developers to query and run analytics against.

The idea is to teach developers how Google technologies like Big Query and Datalab can be used to gather and merge data from different sources, analyze that merged data, then build and train models with TensorFlow to predict movements in financial markets.

Today’s tutorials are, of course, just a clever marketing ploy by Google, as it aims to attract developers to use its machine learning tools. It comes just a few months after Google open-sourced its core TensorFlow machine learning technology that underpins products like Google Translate.

At the time TensorFlow was released to the public, Google CEO Sundar Pichai said the move was designed to help foment more research in the machine learning arena by making it available to developers, academics and hobbyists alike.

Photo Credit: San Diego Shooter via Compfight cc