Google Prediction API graduates from Code Labs with new features

Google’s Prediction API has graduated from the Google Code Labs programme and is now a fully-fledged API hosted on Google Code.

The Prediction API provides businesses and developers with a flexible machine learning in-the-cloud solution to all kinds of problems which, when understood, could enhance a businesses’ success. It can be used for everything from providing recommendations, filtering feeds based on user feedback, automatic customisations, sentiment analysis and more. Users upload their training data to Google Storage, and the service then finds a predictive model which fits the data and that can be used for future predictions, provided on API calls.

Since its launch at Google IO, there have been a number of improvements to the service. More detailed statistics on the model performance is provided to help developers tune their results, which should provide users with more accurate predictions in the long term. The graduation from Code Labs also brings a new version of the API, which supports two major new features. Firstly, PMML import is now supported to an extent (PMML is a data format widely used for transmitting predictive models and their data between systems). There is also an anomaly detection feature, allowing users to review anomalies detected in their training data sets to improve their models.

“With the launch of this version, Prediction is graduating from Google Code Labs, reflecting Google’s commitment to the API’s development and stability” wrote Google’s Zachary Goldberg, Product Manager. “We have been working hard to give every developer access to machine learning in the cloud to build smarter apps.”

You can access the Prediction API and its code samples and documentation through the Code website.

You might also like...



Why not write for us? Or you could submit an event or a user group in your area. Alternatively just tell us what you think!

Our tools

We've got automatic conversion tools to convert C# to VB.NET, VB.NET to C#. Also you can compress javascript and compress css and generate sql connection strings.

“Memory is like an orgasm. It's a lot better if you don't have to fake it.” - Seymour Cray