An information system for knowledge management


01.06.2009 - 31.12.2010


Harald Reiterer

Aim of the Cooperation

The research project is based on a cooperation between the working groupinformation systems of the University of Constance and the department for scientific documentation and information of Byk Gulden, a company in Constance. The project’s goal is the design of an information system for knowledge management; in a shared effort the experts will develop the relevant methods, concepts, and tools. In the company’s department for scientific documentation and information the new system will serve to make research strategies and results available both fast and on a large scale.
The University of Constance is proud to contribute latest research results to this project. Our contribution includes know-how regarding methods and tools for internet research; as well as concerning the administration of research results by means of information systems for knowledge management.

Tasks of knowledge management

For many organisations knowledge has become an essential resource. To be successful, an organisation needs to be able to transform both the individual employee’s and the company’s broader knowledge into organisational knowledge. This is how the existent potential is made available for the company and can be re-used if need be. Despite knowledge’s increasing importance only an estimated 30% of existent organisational knowledge is exploited by companies, the remaining 70% go to waste [Zucker, Schmitz 1994]. Studies reveal also that most companies have difficulties in making use of knowledge; for instance when costlyerrors occur due to unavailable information, or when crucial knowledge ‘leaves’ together with specific employees [Spek, Hoog 1994].
Today, organisational knowledge is usually stored as documents and saved in information systems (e.g. data warehouses). It is possible to exchange these documents between users and/or systems. Examples of documents that record an organisation’s information are: handbooks for various jobs, project reports, or product descriptions. As a consequence documents may contain externalised coded information about different aspects regarding organisational processes and tasks.
The department that is responsible for scientific documentation and information at BykGulden is familiar with the whole context of searchingfor, processing, and distributing knowledge as documents. This is already supported by a variety of information technologies. Therefore the department in question is an ideal partner and at the same time a welcome experimental stage for further research.
There are three levels of knowledge management: the strategic level, the organisational level, and the level of information technology. Specific methods, concepts, and tools are to be developed for each of the three. Besides, the levels influence one another. This is the case, for example, with strategic knowledge planning: it helps to outline how knowledge management might support the company’s objectives and thus identifies the goals to be pursued on the organisational level as well as on that of information technology.

Project Outline/Sketch

In its research work the University of Constance focuses on the level of information technology; special attention is paid to the requirements of strategic knowledge planning, to the specific company as well as to the employees. An information system for knowledge management figures as a platform that supports the exchange of knowledge. As such, knowledge is exchanged as structured information, or so-called explicit knowledge. The higher the information’s degree of structuring, the easier it is to process and recycle explicit knowledge. There are several ways to structure information; an easy and due to the internet familiar one is the use of hyperlinks. 
To date there are no all-inclusive solutions regarding an information system for knowledge management. Instead, the respective tools combine information systems of various fields available at present [Thiesse, Bach 1999]; such as:

  • search engines for the internet
  • systems for groupware and workflow management
  • systems for document management
  • data warehouse/OLAP systems
  • intranet (as technical platform for communication support)

The project’s objective is to design a special system structure Byk Gulden might use for an advanced knowledge management; of particular importance is, one, that the information system integrates the categories listed above and, two, that it takes into account the specific needs of the company. In order to do justice to the project’s research character, it is especially new technologies that are to be considered (e.g. agent based search engines, agent based transfer of knowledge, intranet based communication support).

Apart from developing a new system architecture the project’s other focus is on finding suitable methods to structure knowledge so that the latter is accessible on a larger scale. One possible approach would be to draw on ontology. An ontological approach is based on a mutual understanding about a specific realm of knowledge people exchange sometimes they do so via information systems. Metaphorically speaking, an ontology is a map that depicts the knowledge of an organisation. The better the map, the easier it is to re-locate specific knowledge. The design of ontological knowledge maps is crucial to knowledge management: it is a way to organize and structure knowledge. Examples for ontological approaches are organisation ontology, project ontology, publication ontology etc. Part of our project is to analyze the suitability of ontologies for the structuring of knowledge. 
Another major objective is to identify suitable models for the visualization of knowledge; e.g. knowledge maps, diagrams of knowledge structure, mind maps, semantic networks. For further ideas findings from research done in the field of Information Visualization are to be analysed [Card et al. 1999]. The employment of suitable means of knowledge visualization is to help users to quickly get an overview of contents and structures. This way explorative user behaviour is to be supported (knowledge navigation), and on the other hand it is meant to promote thepossibilities of associatively linking contents (knowledge mining). 
Summing up, there are three major objectives the University of Constance pursues with this project: the development of methods for ontology design; the invention of suitable models for the visualization of knowledge; and, finally, the presentation of a model structure that serves to support the realization of an information system for knowledge management based on the intranet. All three, the methodological ideas, the visualization models, as well as the system architecture, will be realized as prototypes; in a second step, these samples will be evaluated by Byk Gulden’s department of scientific documentation and information.

Expected benefit

There are several expectations to be met when realizing the ideas developed in this project. The company’s expectations center around an increased practical use of information systems that support internal knowledge management:

  • Support of shared use of knowledge:
    The ontological basis serves as a group agreement on the importance of information. Classification enables everybody to easier access documents (explicit knowledge) a specific person added to the information system.
  • Support of knowledge re-use
    Knowledge navigation and knowledge reuse is made easier thanks to ontological classification and structuring of knowledge and the employment of suitable visualizations.
  • Deduction of new facts due to background knowledge
    Both the ontological structure and the employment of relevant visualizations help to better identify and analyze links between documents. This is a way of knowledge mining.
  • Easy integration of heterogeneous sources of information
    Ontological approaches are based on a binding shared terminology. Other sources of information can be transferred to this shared terminology.


  • Card S., Mackinlay J., Shneiderman B.: Readings in Information Visualization, Morgan Kaufmann Publishers, San Francisco, 1999
  • Spek R., de Hoog R.: Towards a methodology for knowledge management, Technical Note Knowledge Management Network, 23. November 1994
  • Thiesse F., Bach V.: Tools und Architekturen für Business Knowledge Management, in: Bach V., Vogler P., Österle H. (Hrsg.): Business Knowledge Managment: Praxiserfahrungen mit Intranet-basierten Lösungen, Springer, Berlin, 1999
  • Zucker B., Schmitz C.: Knowledge Flow Management: Wissen nutzen statt verspielen, in: Gablers Magazin 8(1994) 11-12, S.62-65