Machine Learning Now Available in Google BigQuery

Machine Learning Now Available in Google BigQuery

One of Google Cloud Platform’s most popular products, BigQuery, recently announced the addition of some exciting new functionality: the ability to create and execute machine learning (ML) models directly inside of the platform. BigQuery has become incredibly popular because it enablesinteractive analysis of large datasets, making it easy for businesses to share meaningful insights and develop solutions with analytics. However, often times users of BigQuery aren’t using ML to better understand the data they are generating because they may not have the background needed to apply machine learning techniques.

BigQuery ML is a set of simple SQL language extensions that enables users to utilize popular ML capabilities without advanced ML skills. BigQuery ML sets smart defaults automatically to take care of data transformation, leading to a seamless and easy to use experience. This article will focus on BigQuery’s new ML functionality, but if you’re interested in other new BigQuery features, check out the latest announcements from Google here.

How It Works


Inside of the BigQuery user interface, users now have the ability to create ML models using standard SQL queries. BigQuery ML currently supports the following three types of ML models:

1. Linear regression — These models can be used for predicting a numerical value.
2. Binary logistic regression — These models can be used for predicting one of two classes (such as identifying whether an email is spam).
3. Multiclass logistic regression for classification — These models can be used to predict more than two classes such as whether an input is “low-value”, “medium-value”, or “high-value”.

The model output and metrics can all be found in the BigQuery UI after the query has been processed. For more info on what’s going on behind the scenes, read more here.

For more advanced users, BigQuery ML is also available through BigQuery’s API. In this example, Google walks through a connection between your Datalab instance to a BigQuery dataset and creating an ML model through BigQuery API’s.

How BigQuery Improves Your ML Processes

BigQuery has made ad-hoc analysis amazingly simple with its fully managed service and intuitive UI. As BigQuery now begins to branch into the machine learning space (and others), the ML process will see similar process improvements. With BigQuery ML, data analysts no longer need to move data between environments. The speed and ease to produce ML models becomes significantly more frictionless with BQ ML.

BigQuery ML enables more people to begin engaging with ML. No longer do you need python or other coding experience with BigQuery ML, you simply need to use standard SQL queries.

Ready to get started? Check out Google’s guide for Getting Started with BigQuery ML for Data Analysts. Don’t hesitate to contact us for more information on the Google Cloud Platform.

2018-11-26T13:10:54+00:00November 13, 2018|Categories: Cloud & Machine Learning, Fusion Blog|Tags: , , |

About the Author:

Matt Powers
Matt Powers has spent the last 2 years in a technology delivery role for Maven Wave focusing on machine learning and big data. Matt recently worked at Google as a customer engineer helping organizations take advantage of Google Cloud Platform's computing, big data, and machine learning capabilities.