7.1 Introduction

The ultimate goal in building a BigML Clustering is being able to make predictions for previously unseen instances with an unknown label. Predictions for clusters are referred to as Centroidss, since they aim to identify the nearest Centroids for new data. In BigML, you can make predictions with clusters based on single instances or multiple instances in batch. Each prediction tells you which is the nearest centroid to your input data along with the distance separating them.

The predictions tab in the main menu of your BigML Dashboard is where all of your saved centroids are listed (see Figure 7.1 .)

Cluster predictions are saved under the cluster analysis option in the menu. (See Figure 7.2 .)

From this view you can select to view the list for your single instances centroits or your batch centroids by clicking the corresponding icons. (See Figure 7.3 and Figure 7.4 .)

In the predictions list view, you can see, for each prediction, the Cluster icon used for the prediction, the Name of the prediction, the nearest Centroid predicted, the Distance from the specified input data, and the Age (time since the prediction was created). You can also search your predictions by name by clicking the search icon on the top right menu in Figure 7.1 .

By default, when you first create an account at BigML, or every time that you start a new Project, your list view for predictions will be empty. (See Figure 7.5 .)

Centroids are saved under the cluster icon in the menu options on the top left corner of the predictions list view.