Working with nested JSON strings/files in Julia

No Medium membership? Read this article for free here.

Why and when is JSON used?

Nowadays, there are tons of softwares in all types of scientific and/or business fields that produce/output data expected to be further analysed/manipulated. Data exchange between different platforms/software/languages is prevalent in a data analyst’s daily routine, but it also creates all sorts of issues that can be subsumed under the umbrella of the so-called interoperability problem. This is exactly the problem of finding a common data exchange format between different platforms/software/languages.

Nowadays, JavaScript Object Notation (JSON) is getting more and more popular as the data-exchange format that faces the problem. At least in NLP, a data-intensive field, JSON strings/files are ubiquitous. JSON is a lightweight human-readable text-based serialization format that is easily manipulable, i.e. JSON strings can easily be parsed and generated.

The JSON String and the Goal

The great power of this hierarchical way of representing data is that it allows arbitrarily many layers of nested information. Let’s take a real life scenario of extracting specific values out of deeply nested attributes in a JSON string. For a large scale annotation…