context-aware computing

SUEM: An Ontology For Semantic User Environment Modeling

Project Overview

Advances in mobile and ubiquitous computing, such as context-aware systems, have reshaped the lives of people, by providing them with a wide range of context-based services through various application domains (healthcare, entertainment and leisure, etc.). However, delivering such services requires first building-up a global view of the user’s context. This is done by, first, collecting knowledge about the user and his surrounding environments, and then by modeling and reasoning on collected data pieces while considering the relations that exist among them. Nonetheless, facing (i) the heterogeneity of collected data (i.e., data may have different types and formats), (ii) the heterogeneity of data sources from which data are collected (i.e., IoT sensor networks, social networks, online public databases, etc.), and (iii) the dynamicity of the environment where the user can be located, the knowledge representation process should go beyond classical modeling techniques, to support more comprehensive semantic web techniques with flexible data structure, such as ontologies.

On this basis, we propose in the following a Semantic User Environment Modeling (SUEM) ontology, for modeling collected context data. This ontology is (1) generic, which makes it re-usable in various application domains (e.g., context-dependent privacy risks' inference), (2) modular, an extensible ontology that can be adapted to domain-specific particularities, (3) multi-modal, supports multiple data having different types and format (e.g., scalar data, multimedia data), and (4) multi-source, considers data that can be acquired from different types of data sources deriving from both Connected environments (e.g., IoT sensors networks) and Web environments (e.g., social networks).

The SUEM ontology introduces concepts and relations to represent received knowledge about users, domains of interest, and environments. As shown in the figure below, SUEM is made of three main layers. First, the core layer, comprising elements to represent the generic aspects (i.e., domain-independent aspects) of both users (i.e., user’s personal data) and environments. Then, the pluggable domain-specific user model layer, responsible for integrating user data that are related to domain-specific applications (medical data, social data, financial data, professional data, etc.). And finally, the pluggable domain-specific environment model layer, responsible for aligning the core layer with external ontologies that describe detailed components of specific environments (e.g., building, home, mall, city).

SUEM Ontology Model

Research Areas

  • Computer Science
  • Semantic Web
  • Context-aware Computing
  • Context Modeling

PhD Student

Karam Bou Chaaya

Project Members

  • Richard Chbeir
  • Philippe Arnould
  • Mahmoud Barhamgi
  • Djamal Benslimane


The LIUPPA (laboratory) 

Located in Anglet - France, is a computer science laboratory related to the “Université de PAU et des Pays de l’Adour” (UPPA), and interested in the formalization of digital ecosystems in multimedia sensory networks and microgrids.


  1. We have proposed a new User profile Model that covers the representation of all generic user-personal data from a privacy perspective (i.e., both identifiable and sensitive data), while supporting extensibility to represent domain-specific data.
  2. We have aligned our Environment Model with existing ontologies (e.g., SOSA/SSN, HSSN) that include concepts representing generic aspects of both static and dynamic environments (e.g., concepts expressing static and dynamic systems and devices).
  3. We have introduced new concepts describing shared data items with data consumers, and their sharing characteristics.
  4. We have combined both User and Environment models through inter-entities relations, in order to build-up the core layer of the SUEM ontology. Then, we have introduced two pluggable domain-specific layers to support re-usability in various application domains.

External Links