Back to the previous page

The SemPryv project aims at incorporating a semantization layer to the pryv.io middleware for health data management, providing semantic information inference automatically. The new module will allow Pryv to enhance the findability, search-ability and categroization of its dataset middleware.

sempryv_architecture

SemPryv aims to stream data enrichment by providing semantic annotation capabilities on the Pryv.io middleware. The semantic annotation process associates high-level ontology concepts to the stream events. It can be done in two ways:

  1. Manually by searching well-known ontology providers (such as bioportal), or
  2. Semi-automated where annotation suggestions are provided to the users. These suggestions are derived by predefined rules that experts can modify and save them in the system’s knowledge graph.
  3. Fully automated suggestions: They derived by machine learning models that have been trained on synthetic data from mobile apps combined by users existing annotations.

The architecture of SemPryv is depicted in the picture below. SemPryv has two main components, a web User Interface for end users and experts and a back end that exposes the core services as a REST API to external applications. It also connects to a series of providers for semantic vocabularies and also includes endpoints dedicated for the import/export of HL7 FHIR - compliant data streams, represented as bundle collections of observations. Since the annotations are ready, streams can be exposed to the Pryv again with all of their metadata.


Publications

  • Decentralized semantic provision of personal health streams. J.-P. Calbimonte, O. Aidonopoulos, F. Dubosson, B. Pocklington, I. Kebets, P.-M. Legris & M. Schumacher. Journal of Web Semantics 76, 100774. 2023.  https://doi.org/10.1016/j.websem.2023.100774 
  • Semi-automatic Semantic Enrichment of Personal Data Streams. J.-P. Calbimonte, F. Dubosson, I. Kebets, P.-M. Legris & M. Schumacher. Proceedings of the Posters and Demo Track of the 15th International Conference on Semantic Systems co-located with 15th International Conference on Semantic Systems (SEMANTiCS 2019). Vol-2451, 2019.