How to use NLQ and ChatBot to query data in Pyramid

How to use NLQ and ChatBot features to query data in Pyramid

The Natural language query (NLQ) and ChatBot features in Pyramid have been further enhanced to enable you  harness the collective learnings of other users. The platform tracks prior queries and then surfaces your most recent and most often-used queries throughout the enterprise. This enables you to utilize the power of a collective team within the platform by reusing questions and guiding them to the best questions asked from the team level and also within the data model level.

See in this Video how the NLQ and ChatBot Queries work in Pyramid



Step by Step Guide

Explore these queries using 3 questions a user or an analyst might have.

1. How did each Product category perform this year’s second quarter compared to last year’s second quarter?

a) Type on the chatbot “give me sales for Product Category in a line chart”

You get prompted with the top 5 questions based on most asked question in a team and the community questions within the model.  As you type you will get suggestions on pervious questions and closely related measures and dimensions within the model. Once you enter your question you will be given the possible visual.

Since we are not analyzing by “Channel”  then Type “Remove Channel” in the chatbot .

b) As we are analyzing by quarters we will “add cal data full quarter” by typing in the chatbot.

Based on our visual you can see that Bikes contributes the largest sales and compared to previous 2nd quarter , the last quarter we had the most sales.

2. Analyze based on Sex and what is the most expensive Product category?

a) Type on the chatbot “Add Sex” . Sex is a synonym of gender, since these recommended queries handle words syntax such as thesaurus well, it will look for an alternative word within the data.

b) Now type “which Product Category is the most expensive”. The engine's internal smarts include the ability to "stem" words or find known derivatives of a word to identify entities or other key words and phrases. It found expenses as a closely related word for expensive.

Based on the visual you can see that the most expensive product its Bikes and males are the one who contribute the most compared to females.


3. What are the predicted sales for the next 4 quarters.

a) Type on the chatbot “predict sales for the next 4 quarters”. The chatbot will try and solve the query you have requested, and it will give you the best result based on data and choose an algorithm that best suits the data.

As a user I was able to successfully build a report that analyses product category by sales as well as expenses and predict future expenses-based gender.

4. Restarting the NLQ Chat

Finally, key words such as reset/ restart/ start over can be used to reset the NLQ Chat session and the query or alternatives select reset icon.

The NLQ and ChatBot features are part of the broader AI capabilities available throughout the platform, other AI driven capabilities include Smart Insights, Explain and Chat GPT / Open AI integrations.

if you have questions or suggestions for this article, leave us a comment



Content aside

  • 5 mths agoLast active
  • 58Views