The Hera research program (started at the Technische Universiteit Eindhoven and now a collaboration between research groups from the Technische Universiteit Delft (NL), the Technische Universiteit Eindhoven (NL), the Vrije Universiteit Brussel (B), the University of Namur (B), and the Erasmus University Rotterdam (NL), and some of their fellow researchers) combines research in the area of Web Engineering and Web-based Information Systems (WIS). This research concentrates on
- the design and engineering of WIS
- a design methodology for WIS with automated presentation generation that is adapted to device capabilities and user preferences
- the technology of the Semantic Web, e.g. XML, RDF(S), OWL
- tools to automatically generate (adaptive) hypermedia access to data from online information sources
- data integration and distribution in WIS and on the Web, e.g. wrapping, mediation, distributed querying
- user profiling and the role of user information in personalization and adaptation
- enriched information delivery through data transformation techniques
- Web services, Semantic Web services, and service-oriented computing
Hera is also used as the name of a model-driven design approach and specification framework focusing on the development of context-dependent or personalized WIS. The approach found its origins in an approach for hypermedia presentation generation which gave the first engine complying with this method its name, HPG. The approach distinguishes three main models that specify the generation of hypermedia presentations over available content data: a model for the content, a model for the hypermedia navigation construction, and a model for the presentation construction enable the creation of a hypermedia-based view over the content. Originally, in the first generation of the approach and its HPG toolset, the models specified a transformation from the content to the presentation. The engine that was compliant with this definition was based on XSLT and is therefore also known as HPG-XSLT. One of the characteristic aspects that HPG-XSLT supported was adaptation, for example to produce different presentations from a single design in which the "translation" to formats such as HTML, SMIL, and WML was dealt with generically.
Characteristic for the Hera models is not only their focus on user- and context-adaptation support, but also the choice to base the models on RDF and RDFS. The use of Web standards such as RDF(S) as a modeling paradigm facilitates easy deployment on very heterogeneous data sources: the only assumption made is that a semistructured description of the domain is available for processing. Not only is such a representation less costly to develop than any alternative, but it also enables the reuse of existing knowledge and flexible integration of several separate data sources into a single hypermedia presentation.
During further research into the development of the approach, support was extended for more advanced dynamics. Whereas the first XSLT-based approach primarily transformed the original content data into a hypermedia document, with which the user could interact by following links with a Web browser, one of the next engine versions allowed the inclusion of form processing, which led to the support of other kinds of user interaction while retaining the hypermedia-based nature. Out of this effort, a Java-based version of the engine became available that used RDF queries to specify the data involved in the forms. This version is known as HPG-Java.
The experience from these HPG-based versions and the aim for further exploitation of the RDF-based nature of the models have led to a further refinement of the approach in what is now termed Hera-S. The Hera-S compliant models do combine the original hypermedia-based spirit of the Hera models with more extensive use of RDF querying and storage. Realizing this, in Hera-S RDF data processing using the SeRQL query language caters for extra flexibility and interoperability. Have a look at chapter 10 from Rossi, Pastor, Scwabe and Olsina's "Web Engineering: Modelling and Implementing Web Applications" book published by Springer for a presentation of Hera and Hera-S.
Current Topics and VacanciesSome selected topics we are currently working on:
- distributed user modeling, e.g. in semantics-based systems or in adaptive learning environments
- semantic interoperability
- on-the-fly semantic data integration
- distributed RDF querying
- feature modeling and software variability
- aspect-oriented adaptation
- semantic web services
- time-based reasoning
In these and related topics there are usually positions available for PhD-students. Candidates that are interested can always send us an email.
Hera is now a collaboration between research groups from the Technische Universiteit Delft (NL), the Technische Universiteit Eindhoven (NL), the Vrije Universiteit Brussel (B), the University of Namur (B), and the Erasmus University Rotterdam (NL), and some of their fellow researchers.
People that were or are associated with Hera:
- Lamia Abo Zaid M.Sc.
- dr. Ad Aerts
- dr. ir. Peter Barna
- dr. ir. ing. Pieter Bellekens
- Martin Björkman M.Sc.
- dr. Jeen Broekstra
- Mathieu Cardinael
- dr. Sven Casteleyn
- Jurgen David M.Sc.
- dr.-ing. Zoltán Fiala
- dr. ir. Flavius Frasincar
- prof. dr. ir. Geert-Jan Houben
- ir. Rene Kochen
- ir. ing. Lawrence Mees
- Viorel Milea M.Sc.
- dr. Michael Mrissa
- ir. ing. Bert Okkerse
- drs. Reinier Post
- ir. Bas Rutten
- ir. Martijn Schuijers
- ir. Balpreet Singh
- ir. Kees van der Sluijs
- dr. Philippe Thiran
- dr. ir. Richard Vdovjak
- William van Woensel M.Sc.
Many papers have been published on Hera and on its associated research projects. For an up-to-date selection of papers it is best to visit the publication sections of the Websites of some of the Hera members. For example, have a look for publications on the Website of Geert-Jan Houben (direct link).
In the context of the Hera research a variety of prototype software and demonstrators have been developed. Several versions of a Hera engine to implement Hera models have been realized, e.g. HPG-XSLT, HPG-Java and Hera-S. Also, various auxiliary software - related to the Hera approach - were created like (Hera) model builders, AMACONT-components, etc.