BigML News, Issue #8, May 2014
BigML - Machine learning made easy

 

Hi <<Full Name>>,


BigML is very excited to share with you our Spring 2014 Release. This is a significant update to our platform that introduces a Clustering algorithm, new dataset creation & filtering options, major enhancements for BigMLer and more. Details on the release follow below, and we also encourage you to register for a webinar on June 3 at 10AM Pacific where we’ll showcase this new functionality.
 

The Spring 2014 Release Includes:
 
Cluster Analysis

BigML’s first offering for unsupervised learning allows you to group the most similar instances from your dataset into Clusters. Released in Beta form, BigML’s approach to Clustering is inspired by k-means and features the intuitive workflow and rich visualizations that you’ve come to expect from our service. Once your cluster is built you can predict a centroid (i.e., find the closest centroid for a new data point), and you can also share your clusters with anyone via BigML’s “private link” functionality. Please check out this exciting new addition to our platform and email us with any feedback or suggestions!

 

More Filter and Field Options

The Spring Release brings more of the awesome dataset transformation capabilities of Flatline into the BigML interface. For example, you can now easily filter a dataset using different comparison, equality, missing value, and statistics functions. You can also create new fields discretizing, replacing missing fields, normalizing, and performing all kind of math transformations on previous values of your dataset.


 
Segment-based dataset creation

Have you ever wanted to create a new dataset for further analysis from a specific node in a tree? Now you can! When you’re in a model or sunburst view, simply mouse over a node and then press your keyboard’s shift button. This will freeze the view and allow you to export the rules for that segment and/or create a new dataset with the instances at that node.

 

Dataset exports

Now you can also export datasets from a dataset view into a comma-separated values (.CSV) file. This works very well in combination with the dataset creation above as it can help you identify the instances that follow certain criteria.


 
Ensemble Summary Report

You can now get a Summary Report report for your Ensembles! This is a great way to get a quick summary on what fields have the greatest impact on your predictive outcome–something that can be very illustrative when working with new and/or wide datasets.

 

New BigMLer

BigMLer, our popular command line tool for machine learning, now features powerful new evaluation-guided techniques to support advanced predictive modeling. Specifically, through a new subcommand bigmler analyze you can quickly perform smart feature selection and node threshold selection


 

All of this functionality is available to you today—and works with your pre-existing data sources in addition to any new sources that you establish. Just log into your account or fire up Bigmler and get started today!


 
We love to hear from BigML users—please contact us at info@bigml.com with any feedback or questions.

Our mailing address is:
BigML, Inc
2851 NW 9th Street
Suite D, Conifer Plaza Building
Corvallis, Oregon 97330

Add us to your address book


Copyright © 2014 BigML, Inc, All rights reserved.