PhD proposal - Semantic content negotiation for knowledge interchange among heterogeneous systems

Keywords

Semantic Interoperability, Semantic Web, Linked Data, Content Negotiation

Requirements:

Master in computer Science with good theoretical and practical knowledge of Semantic Web Technologies. Programming skills.

Application

Applications should be submitted by e-mail to antoine.zimmermann@emse.fr and maxime.lefrancois@emse.fr with the reference: PhD_SemConneg_2020

Knowledge of Semantic Web Technologies is mandatory.

The position is available immediately and application evaluation will be continuous until the position is filled. Interested candidates should submit:

Context

Content negotiation on the Web allows a data consumer (client) to tell a data provider (server) what it expects in terms of format, language, encoding, security [Fie14]. In return, the server provides data that meets these expectations when it can, or indicates an alternative, for example, that the same information resource is available in another format. However, even when the client's request is satisfied, this does not mean that the client is able to interpret the data correctly. For example, for the same data format, several forms, structures or schemas may exist. The customer may wish to obtain data that conforms to conventional terminology and has certain logical and structural properties.

In particular, in applications that rely on Semantic Web technologies, such as RDF and OWL, an application could expect graph-based data that conform to a specific ontology [Obr03], or that fit a certain data shape, or that is compatible with a given entailment regime [Gli13], [Zim13]. In environments such as the Web of Things, strong constraints may impose requirements on the server or client side due to processing power, bandwidth, or memory limitations [IERC15].

Objective and challenges to consider

The main challenge is to find out how clients and servers can agree on the expected (client-side) or provided (server-side) content automatically, so without the developer of the client application having to contact the server manager, or read natural language documentation. The objective of this thesis is to : (1) determine what properties the client and the server could agree on to negotiate content beyond its simple syntax; (2) define the mechanism (in terms of protocol and algorithm) allowing the client to announce its expectations and how the server reacts to these requirements; (3) consider making negotiation more flexible by introducing an external service in charge of mediation between client and server (data transformation, inference or validation system); (4) introduce a declarative formalism allowing the server to describe the logical and structural properties of its data (possibly relying on SPARQL 1.1 Service Description [Will13], Thing Description [Kab20], or various forms of content descriptions, e.g., [Thu18]).

Previous and ongoing work has been and are being conducted in extending content negotiation towards more flexibility such as [Hol98], [Sve19] or [Lef18].

References