The evolution of Protégé: an environment for knowledge-based systems development

https://doi.org/10.1016/S1071-5819(02)00127-1Get rights and content

Abstract

The Protégé project has come a long way since Mark Musen first built the Protégé meta-tool for knowledge-based systems in 1987. The original tool was a small application, aimed at building knowledge-acquisition tools for a few specialized programs in medical planning. From this initial tool, the Protégé system has evolved into a durable, extensible platform for knowledge-based systems development and research. The current version, Protégé-2000, can be run on a variety of platforms, supports customized user-interface extensions, incorporates the Open Knowledge-Base Connectivity (OKBC) knowledge model, interacts with standard storage formats such as relational databases, XML, and RDF, and has been used by hundreds of individuals and research groups. In this paper, we follow the evolution of the Protégé project through three distinct re-implementations. We describe our overall methodology, our design decisions, and the lessons we have learned over the duration of the project. We believe that our success is one of infrastructure: Protégé is a flexible, well-supported, and robust development environment. Using Protégé, developers and domain experts can easily build effective knowledge-based systems, and researchers can explore ideas in a variety of knowledge-based domains.

Section snippets

Motivation and protégé timeline

The Protégé system is an environment for knowledge-based systems development that has been evolving for over a decade. Protégé began as a small application designed for a medical domain (protocol-based therapy planning), but has evolved into a much more general-purpose set of tools. More recently, Protégé has developed a world-wide community of users, who themselves are adding to Protégé's capabilities, and directing its further evolution.

The original goal of Protégé was to reduce the

Protégé roots: expert systems and knowledge acquisition in the 1980s

The early 1980s were a heady time for Artificial Intelligence (AI). Expert-systems research had produced some stunning successes (Bachant and McDermott, 1984; Buchanan and Shortliffe, 1984). To many people in the field, it seemed that AI was on the verge of a dramatic breakthrough. Perhaps, Hayes-Roth et al. put it best when they wrote:

Over time, the knowledge engineering field will have an impact on all areas of human activity where knowledge provides the power for solving important problems.

Protégé-I

Protégé-I (Musen (1989a), Musen (1989b)) began as a generalization of the Oncocin/Opal architecture—rather than expecting knowledge engineers to build new knowledge-acquisition tools like Opal for every new domain, the Protégé meta-tool generated a knowledge-acquisition tool (KA-tool) from a set of structural concepts. Thus, we designed Protégé-I to further reduce the load on a knowledge engineer. Fig. 4 shows a flow diagram for the use of Protégé-I, with three classes of users: (1) knowledge

Protégé-II: problem-solving methods and the downhill flow assumption

The most significant difference between the original Protégé and the Protégé-II version was the idea of reusable problem-solving methods. Following Chandrasekaran (1983), Chandrasekaran (1986), Protégé-II allowed developers to build inference mechanisms in an entirely separate component, a problem-solving method, which could be developed independently from the knowledge base. These problem-solving methods (PSMs) were generic algorithms that could be used with different knowledge bases to solve

Protégé/Win: popularizing knowledge-based systems

Protégé-II introduced a number of significant conceptual changes to the basic Protégé idea. In contrast, the development of Protégé/Win was primarily motivated by a pragmatic concern: Protégé-II was built to run only on the NeXTStep operating system, and to expand our user base, we needed to re-implement our system to run under the Windows operating system.3

Protégé-2000: the current implementation

As the Protégé/Win user community grew, and as we received ideas (and feature requests) from this community, we realized it was time to re-engineer the Protégé environment one more time. In contrast to previous iterations, we were motivated neither by the need to drastically change our approach, nor by external forces, such as the need to change hardware or operating system platforms. Instead, we were responding to users’ requests for improving the functionality and generality of the Protégé

Summary and discussion

The four generations of Protégé presented here represent over 16 years of research and system development. Protégé has evolved from a proof of concept and initial prototype to a comprehensive system with an active user community. As we described this evolution, we have highlighted the differences and augmentations from one version to the next. However, there have been some fundamental ideas of knowledge-based systems development that remained unchanged throughout this evolution:

  • Knowledge-based

Acknowledgements

Parts of this work were funded by the High Performance Knowledge Base Project of the Defense Advanced Research Projects Agency (Contract N660001-97-C-8549) and the Space and Naval Warfare Systems Center (contract N6001-94-D-6052). As should be clear, we are greatly indebted to our users, for this project would not have flourished without their active and continued support. In addition to the authors, we acknowledge and thank the many people who have contributed to the development of Protégé.

References (64)

  • K. O’Hara et al.

    Generalised directive modelsintegrating model development and knowledge acquisition

    International Journal of Human–Computer Studies

    (1998)
  • M. Peleg et al.

    Sharable representation of clinical guidelines in GLIFrelationship to the arden syntax

    Journal of Biomedical Informatics

    (2001)
  • A.R. Puerta et al.

    A multiple-method knowledge-acqusition shell for the automatic generation of knowledge-acquisition tools

    Knowledge Acquisition

    (1992)
  • K. Sandahl

    Transferring knowledge from active expert to end-user environment

    Knowledge Acquisition

    (1994)
  • B. Wielinga et al.

    KADSa modeling approach to knowledge engineering

    Knowledge Acquisition

    (1992)
  • J. Bachant et al.

    R1 revisitedfour years in the trenches

    AI Magazine

    (1984)
  • T. Berners-Lee et al.

    The semantic web

    Scientific American

    (2001)
  • J.A. Breuker et al.

    The CommonKADS Library for Expertise Modelling

    (1994)
  • Buchanan, B.G., Shortliffe, E.H. (Eds.), 1984. Rule-Based Expert Systems: The Mycin Experiments of the Stanford...
  • B. Buchanan et al.

    Constructing an expert system

  • B. Chandrasekaran

    Towards a taxonomy of problem-solving types

    AI Magazine

    (1983)
  • B. Chandrasekaran

    Generic tasks in knowledge-based reasoninghigh-level building blocks for expert system design

    IEEE Expert

    (1986)
  • Chaudhri, V.K., Farquhar, A., Fikes, R., Karp, P.D., Rice, J.P., 1998. OKBC: a programmatic foundation for knowledge...
  • Crubezy, M., Lu, W., Motta, E., Musen, M.A., 2001. The internet reasoning service: delivering configurable...
  • Domingue, J., 1998. Tadzebao and WebOnto: Discussing, browsing, and editing ontologies on the web. Proceedings of the...
  • A. Farqahar et al.

    The ontolingua servera tool for collaborative ontology construction

    International Journal of Human–Computer Studies

    (1997)
  • Farqahar, A., Fikes, R., Pratt, W., Rice, J., 1995. Collaborative ontology construction for information integration....
  • Fensel, D., Benjamins, V.R., Motta, E., Wielinga, B., 1999. UPML: A framework for knowledge system reuse. International...
  • Fridsma, D., Gennari, J.H., Musen, M.A., 1996. Making generic guidelines site-specific. Proceedings of the 20th AMIA...
  • Genesereth, M.R., Fikes, R., 1992. Knowledge interchange format, Version 3.0 Reference Manual. Technical Report...
  • Gennari, J., 1993. A brief guide to MAITRE and MODEL: An ontology editor and a frame-based knowledge representation...
  • Gennari, J.H., Altman, R.B., Musen, M.A., 1995. Reuse with Protégé-II: from elevators to ribosomes. Proceedings of the...
  • Cited by (764)

    • Using ontologies for life science text-based resource organization

      2023, Artificial Intelligence in the Life Sciences
    View all citing articles on Scopus
    View full text