State-of-the-Art of User Modeling Ontologies

We describe the state-of-the-art of user modeling ontologies pointing to the domains, advantages, and drawbacks of existing user modeling ontologies.

Note that I still need to look into those user modeling ontologies that you listed. It's just more convenient to do so from the university. --Christoph

  •  ontology (see references)
  • Semantic Web Usage Log Preparation Model (SWULPM): in  Construction of Ontology-Based User Model for Web Personalization: Web personalization to improve access relevant information from the Web. Systems require some representation for user interest models in order to provide a backbone for information recommending and reasoning. This approach draws on concept graphs (c-graphs). Related work is not given.

FOAF

While FOAF itself does not support user modeling, it can be used as a foundation, as it allows for representing users, their accounts on websites, and their relations to other users. --Christoph

Gumo

The  General User Model Ontology (GUMO) aims at a commonly agreed on and uniform interpretation of distributed user models in intelligent semantic web environments. The development of GUMO was influenced by UserML, SUMO, and UbisWord?. It is represented in OWL and focuses on the markup of user models, rather than user modeling, thus, excluding the inference techniques for respective adaptations. The ontology includes concepts such as user's heart beat, age, current position, birthplace, abilities, interest, and very general preferences such as specific wines, sports, or readings.

UbisWord?

 Ubisworld.org is a platform, which allows users to model the intelligent environment as an ontology of concepts such as persons, objects, and locations as well as times, events, and their properties and features.

UMMO (no reuse)

The  User Modeling Meta Ontology (UMMO) (Peter Brusilovsky is involved). The challenge of such meta-ontology construction is not to identify the lower tier concepts that correspond to individual approaches, but to work out and verbalize the upper tier concepts that would help generalize about UM methodology. Consequently, UMMO can help us to learn about the state-of-art in user modeling, but doesn't provide a domain-specific specification of user characteristics (as we are aiming at). But we can try to position us in UMMO, which will allow a better overview on related work (i.e. the approaches that are close to ours in the ontology).

UserML

The  User Modeling Language (UserML) was originally developed for the area of ubiquitous computing, which aims as model of human-computer interaction in which information processing is integrated into everyday objects and activities as opposed to desktop scenarios, in which users only engage a single device.

SUMO

The  Suggested Upper Merged Ontology (SUMO), a formal upper ontology, is extremely broad in scope and is intended to serve as a semantic foundation for search, interoperation, and communication in the semantic web. (see also  wikipedia)

I'm not sure whether upper model ontologies are of interest here. At least not for the user modeling itself, but certainly for tagging, and for expressing what topics a user is interested in. For that they are good because they contain a huge supply of real-world concepts. Then, probably  UMBEL is a better choice, as it is closely integrated with the (semantic) web, e.g. contains lots of links from general "upper ontology" concepts to concrete semantic web ontologies (e.g. FOAF), and to places where you find realistic descriptions of these concepts (e.g. Wikipedia). --Christoph