This paper is intended to serve as a comprehensive introduction to the emerging eld concerned with the design and use of ontologies. We observe that disparate backgrounds, languages, tools, and techniques are a major barrier to e ective communication among people, organisations, and/or software systems. We showhowthe development and implementation of an explicit account of a shared understanding (i.e. an `ontology') in a given subject area, can improve such communication, which in turn, can give rise to greater reuse and sharing, inter-operability, and more reliable software. After motivating their need, we clarify just what ontologies are and what purposes they serve. We outline a methodology for developing and evaluating ontologies, rst discussing informal techniques, concerning such issues as scoping, handling ambiguity, reaching agreement and producing de nitions. We then consider the bene ts of and describe, a more formal approach. We re-visit the scoping phase, and discuss the role of formal languages and techniques in the speci cation, implementation and evaluation of ontologies. Finally, we review the state of the art and practice in this emerging eld,

We outline some requirements for a comprehensive methodology for building ontologies, and review some important work that has been done in the area which could contribute to this goal. We describe our own experiences in constructing a significant ontology, emphasising the ontology capture phase. We first consider the very general issue of categorisation in modelling, and relate it to the process of ontology capture. We then describe the procedure that we used to identify the terms and produce definitions. We describe a successful way to handle ambiguous terms, which can be an enormous obstacle to reaching a shared understanding. Other important findings include: it may not be necessary to identify competency questions before building the ontology; the meta-ontology can be chosen after detailed text definitions are produced; defining terms which are 'cognitively basic' first can lead to less re-work. AIAI-TR-183 Page 1 of 13 1 Introduction Currently, a considerable body of experience...

The use and importance of ontologies is becoming more widespread, however building ontologies is largely a black art. The aim of this paper is to identify and characterise what we currently know and to move towards the longer term goal of developing a comprehensive unified methodology. We first identify dimensions for characterising ontologies, to be used as a basis for noting which techniques and guidelines for building ontologies apply in different circumstances. We then give an overview of the current state of the art, noting that most work addresses just a small part of the life cycle. The very few more complete methods are limited to case studies involving single ontologies and they are hard to compare. In the main part of this paper, we examine two such methods and give a framework for comparing and unifying them. We emphasise that different approaches are required for difference circumstances, and give some guidelines for when to use which techniques. We conclude by ...

In view of the need to provide tools to facilitate the reuse of existing knowledge structures such as ontologies, we present in this paper a system, AKTiveRank, for the ranking of ontologies. AKTiveRank uses as input the search terms provided by a knowledge engineer and, using the output of an ontology search engine, ranks the ontologies. We apply a number of classical metrics in an attempt to investigate their appropriateness for ranking ontologies, and compare the results with a questionnaire-based human study. Our results show that AKTiveRank will have great utility although there is potential for improvement.

One of the main hurdles towards a wide endorsement of ontologies is the high cost of constructing them. Reuse of existing ontologies offers a much cheaper alternative than building new ones from scratch, yet tools to support such reuse are still in their infancy. However, more ontologies are becoming available on the web, and online libraries for storing and indexing ontologies are increasing in number and demand. Search engines have also started to appear, to facilitate search and retrieval of online ontologies. This paper presents a fresh view on constructing ontologies automatically, by identifying, ranking, and merging fragments of online ontologies.

Abstract. Governments often hold very rich data and whilst much of this information is published and available for re-use by others, it is often trapped by poor data structures, locked up in legacy data formats or in fragmented databases. One of the great benefits that Semantic Web (SW) technology offers is facilitating the large scale integration and sharing of distributed data sources. At the heart of information policy in the UK, the Office of Public Sector Information (OPSI) is the part of the UK government charged with enabling the greater re-use of public sector information. This paper describes the actions, findings, and lessons learnt from a pilot study, involving several parts of government and the public sector. The aim was to show to government how they can adopt SW technology for the dissemination, sharing and use of its data. 1

: Many KA research groups are currently focused on the reuse and sharing of ontologies as a means to alleviate the KA bottleneck. Such endeavours often take for granted that a shared view already exists and thereby avoid the problem of reconciling differences. A fundamental problem associated with reuse and sharing is that the underlying ontological and terminological assumptions have not been made sufficiently explicit or agreed upon. The approach suggested in this paper is to allow experts to develop their own knowledge base using a simple knowledge acquisition and representation technique known as Ripple Down Rules and then to apply our reconciliation process which relies on Formal Concept Analysis to build a shared knowledge base. A general framework for this approach is described together with an illustration using the SISYPHUS III knowledge acquisition material. 1. Sharing Knowledge -- the Holy Grail or a Pipe Dream ? Many knowledge acquisition (KA) research groups are currentl...

: When modelling knowledge of multiple experts, it is interesting to build a common expertise model corresponding to the kernel of knowledge common to the experts. Therefore the expertise conflicts between the expertise models of the different experts must be tackled. The domain level of an expertise model can be described through concepts linked by relations, and represented through Sowa's conceptual graph formalism. This report presents a method for conflict management during knowledge modelling from multiple experts: this method relies on the comparison and integration of several conceptual graphs corresponding to different viewpoints, the integration being guided by different integration strategies. For the comparison of the conceptual graphs, we define and exploit possible relations that can link such graphs. The appendix presents a base of conceptual graphs obtained by modelling the knowledge of several experts in traffic accident analysis. Key-words: knowledge acquisition from ...

This paper describes an approach that begins with asynchronously capturing individual conceptual models into separate knowledge-based systems using a technique known as Ripple-Down Rules. The rules are combined and used by Formal Concept Analysis to develop the group's conceptual model thereby revealing common ground and differences between the stakeholders. As group process is offered which allows conflicts to be identified and resolved, where possible, using our negotiation strategies and resolution operators. The individual models are updated based on the negotiations and the cycle repeats. The end result over a number of iterations is a shared conceptual model. Keywords AL01 Knowledge representation; AL04 Knowledge acquisition; HB26 Simulation and modelling IS; Formal Concept Analysis; Group Decision Support; Ripple Down Rules 1. INTRODUCTION The motivation for this work is based on research into acquiring knowledge into knowledgebased systems (KBS 1 ). It has become well acc...

images, données, connaissances apport de recherche