Flow can be used to consume any JSON based data source into a tabular dataset for analysis. It does so without requiring knowledge ahead of time about the structure or schema of the data. This functionality eliminates a significant amount of time when dealing with the acquisition and curation of data from JSON based sources.
In this blog post, I will demonstrate how to integrate, transform, and analyze JSON data.
The example we will explore will focus on the web-based JSON source available here.
The supplied data source contains information on historical Nobel Prize winners. We will learn how to target this web-based source and consume it into a dataset with a single step.
Once we import the data into Flow, we will see a quick example of how to build hypercubes against the data to generate some summary views.
We will then see how to use the expression builder to compute new data points on the fly. We will learn how to slice the generic set and perform a language analysis, highlighting the ability to execute any required transformations on the data.
Finally, we will learn how to export the flattened data to various formats allowing us to persist the output for use elsewhere.
The technique outlined in this example is useful for any data analyst or data scientist who needs to work with JSON sources.
If you do not have a Flow account - register here. Flow is free for personal use.
The video below provides the full worked example of how to import and analyze JSON data. Check out the video here: