My talk this week will be split into two parts. In the first, I shall describe some of my previous research in the field of engineering design, and, in particular, into the automation of the design synthesis task. Since, in general, the acquisition of design synthesis knowledge using standard KA methods is difficult, my work involved investigating the use of previous designs as an alternative source of this knowledge. I will discuss some of the results of my attempts to exploit information of this sort using Machine Learning and Case-Based Reasoning.
In the second part of the presentation, I will talk about some of my related work on the AKT project. One of the principal themes of the AKT work at Edinburgh is that of the 'knowledge audit' - the appraisal of an organisation's knowledge resources, with the aim of providing a better understanding of the nature and extent of these resources, and hence, allowing this knowledge to be managed more effectively. Since, an organisation's databases may be considered to be 'potential' knowledge - their exploitation using, say, ML or CBR could turn them into actual knowledge - there is a case for including databases within the scope of a knowledge audit. I will discuss the feasibility of auditing databases for this purpose.