Master theses

Linked data is prevalent in many emerging technologies: many privacy-oriented and decentralised solutions, such as Solid and Personal Data Spaces, use RDF to represent their data, as it allows for an ecosystem where interopability thrives. Naturally, being able to process this data in the form of querying is important. SPARQL is a query language designed to query such data, allowing applications to use semantic relationships and operators to obtain the data they need. Implementations of the SPARQL specification, often referred to as SPARQL engines, are typically focussed on server and web scale use cases. Some of these implementations have been expanded to consume data streams, such as time series, creating an extension of the SPARQL specification with data stream processing functionality. No implementation of such extensions exist on mobile platforms, even though these data streams are just as common in a mobile context.

In this thesis, you will be researching the feasability of integrating data stream querying support for an existing mobile SPARQL engine. The existing engine supports incremental evaluation on data, with no support for the extended specifications for data streams. Instead, your implementation becomes responsible for handling these specification extensions, converting the raw text input into a valid and usable query, and evaluating data streams using the operators according to the specification. As mobile platforms often limit the memory footprint of applications, these constraints have to be considered whilst analysing the feasability and implementing the stream processing logic.

Alternatively, different ways of visualising regular queries, data streams and stream queries can be researched. Semantic relations and information can be analysed to offer relevant visualisations, such as a specialised time series representation when querying time series data streams, through a concrete mobile or web implementation.