Intuitively discovering and exploring a Wine dataset with the Dynamic Dashboards

Published on 16 January 2018 in Version 4 - 2 minutes read - Last modified on 06 March 2021

Dashboards are an easy way to quickly explore a new dataset without typing any SQL. It usually complements the analyses before or after the querying activity in the SQL Editor. In this scenario we are using Apache Solr as the analytic database backend.

Goal: we are picking a Wine dataset from the Machine Learning Competition website Kaggle and want to get a feeling of the industry and select some good wines similar to one of our best ranked favorite Pink / “Rosé” wine. Let's continue in the video demo:

Importing

Any CSV file can be dragged & dropped and ingested into an index in a few clicks via the Data Import Wizard [link]. The indexed data is immediately queryable and its facets/dimensions will be very fast to explore.

Browsing

The Collection browser got polished in the last releases and provide more information on the columns. The left metadata assist of Hue 4 makes it handy to list them and peak at their content via the sample popup.

Querying

The search box support live prefix filtering of field data and comes with a Solr syntax autocomplete in order to make the querying intuitive and quick. Any field can be inspected for its top values of statistic. This analysis happens very fast as the data is indexed.

On top of this, the Solr 7 Analytic Facets are close to be fully supported in the Dashboard. More on this is coming in a follow-up blog post. The “More like This” feature lets you selected fields you would like to use to find similar records. This is a great way to find similar items like wine in this tutorial, or similar issues, customers, people…

 

Note:

If you are not getting any suggestion, and opening the field information popup on the right assist shows the error below, it means the collection needs to have the Solr Term Handler configured.



HTTP Status 404 - /solr/jira_search/terms

There are no terms to be shown

Coming up Next!

A deeper presentation of Solr's Analytic Facet is coming as well as how we could have Geo-IPed our wine list and visualize it via maps. A revamp of the layout to make the Dashboards even more intuitive and how the SQL Editor and Dashboards are getting tied together will also be on the menu. We wish you some Happy Explorations!


comments powered by Disqus

More recent stories

03 May 2023
Discover the power of Apache Ozone using the Hue File Browser
Read More
23 January 2023
Hue 4.11 and its new dialects and features are out!
Read More