Emerald Group Publishing Limited
Copyright © 2004, Emerald Group Publishing Limited
The word “ontology” is borrowed from philosophy and has been given a special meaning in AI, where it is used to denote a facility for the formal representation of relations among stored items of information, usually by a set of inference rules and embedded in a special language. As the author of this book puts it: “Ontologies define real‐world semantics, which make it possible to link machine‐processable content with meaning for humans based on consensual terminologies”. Ontologies are the vital component (or colloquially, “silver bullet”) of the “semantic Web” visualised by Tim Berners‐Lee and discussed in an “Internet commentary” in Kybernetes Vol. 32, Nos 7/8, pp. 1178‐81, 2003. (Berners‐Lee and Miller, 2002; Berners‐Lee et al., 2001; Euzenat 2002).
The developments are essential to let computers interact with humans as in Star Trek, and it is claimed that (at the present time, here on earth) the computer and telecommunications revolution has so far barely begun, despite the great changes that have already become commonplace.
Some of the possibilities have been indicated in scenarios, such as one by Euzenat in which a software “agent” makes plans and arrangements for a user's trip to attend a particular conference, including travel, hotel and entertainment bookings, with checks on availability and time compatibility and conformity with the attendee's predilections and budget. It is even visualised that the “agent” could be asked to check whether a particular friend from another establishment plan also to attend, and if so to try to arrange to meet for dinner. The “agent” has to interact suitably with hotels and travel companies and with the “agent” of the friend. An ontology is the means of allowing appropriate communication.
Apart from this kind of “personal assistant” function, the semantic Web would allow many new forms of data searching that would go far beyond the capabilities of existing search engines because of the attention to semantics. Several classes of ontologies are described and include versions developed to operate within particular domains of knowledge.
The best‐known languages embodying characteristics of ontologies are XML (Extensible Markup Language) and RDF (Resource Description Framework), along with many others that are reviewed in the book. It is mentioned that the original intention in devising HTML was that it should represent the structure of a document rather than just its layout, but this has not been realised and therefore these further developments are required.
Software tools are reviewed, including ontology editors and systems for data searching and knowledge manipulation using ontologies. Application areas are discussed with the observation that one commercial system, dealing with selling of software, is already up and running.
There have been previous publications dealing with the topic, mainly conference proceedings, but this is believed to be the first single‐author work reviewing the field. It is clear, comprehensive and practical and will certainly be a standard work of reference in this exciting area leading to an entirely new relationship between people and computers.
Berners‐Lee, T. and Miller, E. (2002), “The semantic Web lifts off”, ERCIM News, No. 51, pp. 9‐11, October.
Berners‐Lee, T., Hendler, J. and Lassila, O. (2001), “The semantic Web”, Scientific American, Vol. 284 No. 5, pp. 35‐43, The paper can be downloaded free of charge from: www.scientificamerican.com/2001/0501issue/0501berners‐lee.html May 17.
Euzenat, J. (2002), “A few words about the semantic Web and its development in the ERCIM institutes”, ERCIM News, No. 51, pp. 7‐8, October.