What Is It?

Explore's Natural Language Query (NLQ) Assistant empowers users to simply and consistently extract insights from their data by typing short natural language requests.

Why Is This Important?

NLQ further broadens the audience who can take take advantage of Explore's already easy-to-use AI tools. Less technical users and new users alike can perform powerful analysis by using simple commands like “What drives ‘Sales’?” or by taking advantage of our Suggested Queries. Extended Queries can be shared easily and they are a great way to reproduce the same complex visualization on datasets that are updated frequently. Additional syntax and industry-specific language can be customized through use of Keywords.

For example, a user can ask “Find outliers for ‘Sales’” or “Filter ‘City’ is ‘New York’ or ‘Los Angeles’" (without quotation marks) and the plot will update without having to use any other parts of the interface.

How? 

In addition to posing your own questions, you can choose to use Suggested Queries and set up Extended Queries to build out workflow triggers for non-technical users.

Suggested Queries

Click into the NLQ Assistant bar and a few Suggested Queries will be displayed below. Suggested Queries are recommended AI routines and plots that will most likely reveal statistically interesting patterns and relevant insights for your dataset. 

Keywords

For more advanced users who wish to tailor Suggested Queries more to your specific data, we have included a Keywords file which can be found in the following locations on different platforms:

• Windows: C:\Users\[username]\Documents\Virtualitics\Settings\keyword-bank.csv
• Mac: /Users/[username]/Virtualitics/Settings/keyword-bank.csv
• Linux: [.tar file path]/Virtualitics.[version]/Virtualitics_Data/StreamingAssets/NLQSettings/keyword-bank.csv

This file can be edited to ensure that features with the desired keywords are displayed in the Suggested Queries. This can be very useful for speeding up analysis of new datasets when you know which features are most important to your analysis. Below is an image of a few Keywords from the file:

Extended Queries

In order to make NLQ even more accessible, Explore includes an editable collection of template queries that allow you to expand the phrasings and terminology that the NLQ Assistant will recognize. Extended Queries speed up common or lengthy queries with streamlined phrases, and advanced users in an organization can add more accessible queries to share with others.

The Extended Queries file can be found in the following locations on different platforms:

• Windows: C:\Users\[username]\Documents\Virtualitics\Settings\standard-extended-queries.csv
• Mac: /Users/[username]/Virtualitics/Settings/standard-extended-queries.csv
• Linux: [.tar file path]/Virtualitics.[version]/Virtualitics_Data/StreamingAssets/NLQSettings/standard-extended-queries.csv

The structure of the file is a list of input templates, output templates and optional comments. Users can add custom inputs and queries into the file to create their own Extended Queries, which can include either placeholder feature names like {FEATURE1}, surrounded by curly brackets, or specific feature names like ‘Days_Delayed’ or ‘Sales/SqFt’, surrounded by single quotes. These Extended Queries can greatly simplify repeated analysis. Below is an image of a few Extended Queries from the file: