home profile publications research teaching service awards

Semantic Web Engineering

"The Semantic Web is an extension of the current web in which information is given well-defined meaning, better enabling computers and people to work in cooperation" (Tim Berners-Lee et al). Following the explosive initial growth of the Web, we are faced now with a vast universe of more or less organized information resources.

Databases play a crucial role in this universe. Beyond the billion of static Web pages indexed by search engines, there is a "Deep Web" of database contents that is not indexed due to its dynamically generated responses. It is estimated that the Deep Web exceeds the commonly defined World-Wide Web by a factor of 500 in data volume (source: www.completeplanet.com).

The course is subdivided into four parts. After a brief introduction, case studies are investigated, with emphasis on data intensive scientific applications, in particular Grid and geo services. Next, Web Services and the Semantic Web are introduced by inspecting the pertaining standards. The third part is dedicated to selected database concepts and techniques which contribute to the quest of efficiently adding more semantics to the Web.

Goal is to make students familiar with the state of the art in Web-enabled information systems so that they will be successful database/Internet professionals in IT industry or, alternatively, have a sound knowledge base to specialize towards a scientific career in the field.


see the course mailing list:



Projects are embedded into our work on geo service standardization and the implementation of these standards. In particular, this semester they are tied to the just started European research project EarthServer where institutions like Plymouth Marine Laboratory and NASA collaborate under our guidance.

Specifications heavily make use of UML and XML with XML Schema, Java (servlets), JavaScript, AJAX, etc., in combination with further appropriate tools.
A core of tasks is common to all projects: Understanding the specifications sufficiently, implement, document, and demo them. Demonstration data will be provided.


  • A Semantic Web User Interface Toolkit for Multi-Dimensional Raster Services
    In our work on multi-dimensional raster services we have established flexible, effcient server technology for n-d multi-Terabyte raster data. What we (and our user communities) still lack is versatile client toolkits. In a previous lab work, such a toolkit has been established. While only requiring writing HTML it allows to configure interactive retrieval pages with result data display.
    Task on hand is to enhance this raster service toolkit, apply it to the demos at EarthLook, and make it ready for use by our international partners. To this end, you will get in contact with the EarthServer consortium and collaborate with these international partners.

  • A Semantic Coordinate System Resolver
    The Open Geospatial Consortium (OGC) is a standardization body issuing specifications for geo service interfaces. Our research group regularly contributes specifications which often become adopted standards, such as WCS and WCPS. Currently we are elaborating a specification which will allow to compose Coordinate Reference Systems dynamically. This is needed, among others, to allow retrieval from more-than-2D data, such as 3D image time series and 4D climate/ocean data.
    Task on hand is to implement a resolver for Coordinate Reference System (CRS) identifiers, based on an existing specification draft. The service receives a URL, possibly with GET parameters, and returns an XML definition of the CRS addressed. The resulting service will be installed on the OGC website for worldwide use by companies, governments, and researchers.

  • The Visible Human as a Web Service
    The Visible Human is a 3-D tomogram of a human body available for research in the life sciences. Many services have been established using this data set, including this one. Goal of this project is to import these data into our rasdaman database and offer it via an interactive Web service in the spirit of EarthLook to show feasiblity of the WCPS geo service standard for life science applications. Notably, functionality should go distinctly beyond mere slicing along the x/y/z axis. By inspecting other services it is expected that new, advanced functions will be developed as queries, such as slicing in any angle, and implemented using current Web technology like JavaScript.
  • Integrating a Visual and Semantic Geo Web Service
    The Open Geospatial Consortium is the geo service standardization body. Their most implemented standard is the Web Map Service (WMS) which allows to browse 2D maps in a Google/Bing style. Further standards include feature (roughly: vector) and coverage (roughly: raster) services, the latter being Web Coverage Service for simple extraction and download and Web Coverage Processing Service for advanced processing. The open source version of rasdaman currently supports WCS and WCPS, the commercial version WMS.
    Task is to take the WMS code of the commercial version and integrate it with the open-source code, ie: make the rasdaman WMS interface open source. This involves integration of the metadata on database level, and integration of the Java code into the petascope servlet, the rasdaman OGC geo Web service layer. The resulting test has to be tested thoroughly and on real-life data, it will become part of EarthLook and will be heavily used in the EarthServer project as well; if desired, integration into and active participation in this project is possible.

Further topics can be negotiated - contact me if you have a cool idea!

A project consists of several phases (due dates announced during first class unit):

  • phase 1: source material evaluation
    deliverable: report (~5 pages)
  • phase 2: specification
    deliverable: specification document; should normally include the sections: overview; service specification (including justification of features selected, in case of a subset); architecture; interfaces; demonstration scenarios (including sample requests).
  • phase 3: implementation
    deliverable: documented code, ready for installation at some other place; individual handover appointments will be arranged.
  • phase 4: presentation and final report
    deliverable: presentation in class (~20min), including live demonstration, and final report (will be based on the specification document, but extend it with a description and discussion of the results obtained).

Individual deviations might be possible, due to the individuality of the projects. If you want to deviate then agree with your supervisor until the end of phase 1 at the latest.


...will be done as announced in class, see 1st slide set.



  • [Semantic] Web Service Conferences:
    • Liang-Jie Zhang, Mario Jeckle (eds.): Web services. European conference, ECOWS 2004, Erfurt, Germany, September 27-30, 2004, LNCS 3250, Springer, 2004.
    • Sheila A. McIlraith, Dimitris Plexousakis, Frank van Harmelen (eds.): The Semantic Web--ISWC 2004. Third International Semantic Web Conference, Hiroshima, Japan, November 7-11, 2004, LNCS 3298, Springer-Verlag, 2004.
    • Christoph Bussler [et al.] (eds.)The Semantic Web : research and applications. First European Semantic Web Symposium, ESWS 2004, Heraklion, Crete, Greece, May 10-12, 2004, LNCS 3053, Springer, 2004.
    • Christoph Bussler [et al.] (eds.): Web information systems. WISE 2004 International Workshops, Brisbane, Australia, November 22-24, 2004, LNCS 3307, Springer, 2004.
  • Web Database Conferences
  • Grid Service Conferences:
    • Mario Jeckle, Ryszard Kowalczyk, Peter Braun (eds.): Grid services engineering and management. first international conference, GSEM 2004, Erfurt, Germany, September 27-30, 2004, LNCS 3270, Springer, 2004.
    • Mokrane Bouzeghoub ... [et al.] (eds.): Semantics of a networked world. Semantics for grid databases, first international IFIP conference, ICSNW 2004, Paris, France, June 17-19, 2004, LNCS 3226, Springer-Verlag, 2004.

More Links

  • Web Services Implementation
  • Web Services Implementation Languages
    Given the large variety of Web app implementation languages available it is no surprise that there is hot discussions on which platform is the best one. One line of arguments, which I share to a large extent (but not completely) is presented by two webcasts of a NASA/JPL guy (webcast1, webcast2).
Copyright © 2004-2009 Peter Baumann -- -- tel. +49-173-583 7882 -- Disclaimer