Forward by Professor Nigel Shadbolt |
|
In the 1970s, AI practitioners came to believe that the secret to building effective software programs was to fill them with knowledge. However, they knew very little about the techniques and methods required to extract knowledge either from textural sources, databases or human experts. The field of knowledge acquisition and elicitation was born as a distinct area. I well remember the earliest conferences and workshops on the topic. They were a heady mix of disciplines – psychologists, sociologists and computer scientists. The one unifying observation was that there was a significant bottleneck, i.e. that knowledge acquisition and elicitation was costly and difficult to do and there was little advice and few tools or techniques available. As academics are wont to do in a new topic area, workshops flourished, conferences were born and research projects were funded. Significant progress was made by developing principles, methods and software tools to help extract, organise, validate and implement knowledge. A substantial problem, however, was that the area tended to become an academic enclave and too little attention was given to the actual process of engaging with end-users, application context and the business requirements. Despite 30 years of investment in this area, it is still the case that there is little practical advice and guidance for those embarking on an attempt to organise and regiment knowledge from whatever source. During the 1980s, the discipline known as knowledge acquisition rapidly attracted the attention of not just those interested in building computer systems. The whole area of management science and organizational psychology was in need of techniques for building knowledge structures. In the late 80s and early 90s, the field of knowledge management came to prominence in a number of seminal books. Companies became increasingly interested in their knowledge assets: how they could represent the knowledge held in their companies, how they could protect it and how they could assign value to it. The whole field of knowledge management added a new dimension to the requirement for tools and techniques to formally model knowledge. More recently, the extraordinary emergence of the largest information construct in human history, the World Wide Web, has presented us with a new way of thinking about how we might organise and regiment information. The challenge here is that a wealth of information has to be acquired, retrieved, indexed, sorted, structured and managed. So in the last 30 years, we have seen at least 3 major drivers all of which point to a requirement for knowledge acquisition techniques. To date there has been no single treatment of the practical methods to build systems, in the most general sense, that embody knowledge. Whether these are for the purposes of building decision support systems, organizing a corporate knowledge management programme, putting up sustained intranets or modelling the content of the web. As someone who was involved in the earliest days of the research, it gives me great pleasure to write this foreword. Nick Milton worked with me at the University of Nottingham in the early days of seeking to widen the applicability of knowledge acquisition tools and techniques. He has gained huge experience in the applied and practitioner aspects – how one teaches this material, how you convey the important concepts – and has engineered a wide variety of successful solutions for many clients. This book presents a systematic presentation of processes, procedures and routines to organise a general knowledge acquisition project. The beauty of the approach is that it is independent of whether the project is to deliver a decision support system, a knowledge management product or a website. The approach also allows the incorporation of particular or bespoke in-house methods at various stages in the acquisition process. As someone who on many occasions presented courses, lectures and seminars in the area, it was always somewhat embarrassing to be asked “Where is the definitive text book on the art and craft of knowledge acquisition?”. This book squarely meets that requirement. |
|
Professor Nigel Shadbolt |