Master theses

Writing a highly efficient JSON-LD parser for specific frames

Keywords: Linked Data, RDF, Webdevelopment, JavaScript, Parsers, Parsing

Supervision: Ruben Verborgh Ruben Taelman

Students: max 1

JSON-LD is a subset of JSON that enables semantic descriptions of things. Today, a variety of generic JSON-LD parsers exist in various programming language. The goal of this thesis is to design and implement a meta-parser in JavaScript that can generate parsing code based on a specific JSON-LD frame, such that data adhering to this frame can be parsed in a highly efficient manner.

The aim is that this meta-parser produces parsers that are more efficient compared to using a generic JSON-LD parser.

In order to test this implementation, the student will test this with existing data from schema.org.

However, a lot of JSON-LD documents exist in very specific shapes, for which generic parsers cause too much overhead.

JSON-LD is widely used for representing information on Web pages, which is for example used by search engines such as Google to build up their knowledge graph in order to improve search results.