Virtualitics offers a network extractor tool to create network data from nearly any type of dataset you would normally load into our software. This allows you to take advantage of our network visualizations and analysis tools to better understand your data. Our in-app Network Extractor has three input sections for you to create your network:
- Node Feature: Input a feature that you would like to use to specify the nodes in your network. This should be a column with somewhere between a few hundred and several thousand unique values in order to produce a useful network.
- Pivot Type: Select the type of pivot you would like to perform on the original dataset’s numerical columns when creating your network. The results of the pivot will turn into metadata columns for the nodes.
- Associative Features: Choose the features that you would like to use to create associations between the nodes, which will turn into edges in the network. You can also click on Input All to include all the features. Hovering on a feature and clicking on the X will let you remove it from the list, and you can remove all features with many missing values by clicking on the “Remove sparse features” button next to Input All.
Please note that for large datasets, running the network extractor could take a few minutes, and may cause Virtualitics to close unexpectedly.
Please also refer to our Example Notebooks page for an example of how to use the network extractor through our Python API.