Towards Very Large Knowledge Bases

In the early days of artificial intelligence it was widely believed that powerful computers would, in the future, enable mankind to solve many real-world problems through the use of very general inference procedures and very little domain-specific knowledge. With the benefit of hindsight, this view can now be called quite naive. The field of expert systems, which developed during the early 1970s, embraced the paradigm that Knowledge is Power - even very fast computers require very large amounts of very specific knowledge to solve non-trivial problems. Thus, the field of large knowledge bases has emerged. This book presents progress on building and sharing very large-scale knowledge bases. Progress has been made in specific scientific domains, including molecular biology, where large knowledge bases have become important tools for researchers. Another development is the attention being paid to structuring large knowledge bases. The use of a carefully developed set of concepts, called an ontalogy , is becoming almost standard practice. This text provides a guide to the current state of the art in building and sharing very large knowledge bases, and is intended to act as a catalyst to future research, development and applications.