Besides simple metrics, we often like to focus on percentiles and histograms for data science queries. This video shows how Elasticsearch supports both percentile queries and histogram results.
- [Instructor] A common thing in data science…is to look at percentiles and histograms,…and Elasticsearch actually supports…these queries very well.…So let's take a look at these two types of analysis now.…First, let's take a look at percentiles for bank balances.…I'll start out with get then do bank account search,…just like before.…I'll give it a size of zero, 'cause I just want…the data being returned, now I'll add my aggregations,…and I'm going to give it a name, percent balances.…
For the ag type is where I add percentiles,…and notice it populated this big long list here,…so I have to give it the field mainly,…and that's going to be balance, and it automatically…defined here what intervals I want the percentiles for.…Now I can punch in whatever intervals I want,…but I'll just leave these here for now…to show you what the results look like.…So on the right, you can see we have our aggregations…and the values being returned,…so these are the accounts that are at this different…percentile, this varying percentiles of our balance field.…
In this course, join Ben Sullins as he dives into the inner workings of Elasticsearch combined with Kibana. Ben provides an overview of the architecture, and then goes over the different deployment methods, and how to best structure your data. From there, he demonstrates how to query data, and how to work with Kibana to present your insights.
- Reviewing key Elasticsearch concepts
- Running Elasticsearch in the cloud and locally
- Bulk loading data
- Setting up mappings of data types
- Querying data
- Running basic aggregations
- Creating visualizations and dashboards in Kibana