Case base adaptation using solution-space metrics

In this paper we propose a generalisation of the k-nearest neighbour (k-NN) retrieval method based on an error function using distance metrics in the solution and problem space. It is an interpolative method which is proposed to be effective for sparse case bases. The method applies equally to nominal, continuous and mixed domains, and does not depend upon an embedding n-dimensional space. In continuous Euclidean problem domains, the method is shown to be a generalisation of the Shepard's Interpolation method. We term the retrieval algorithm the Generalised Shepard Nearest Neighbour (GSNN) method. A novel aspect of GSNN is that it provides a general method for interpolation over nominal solution domains. The performance of the retrieval method is examined with reference to the Iris classification problem,and to a simulated sparse nominal value test problem. The introducion of a solution-space metric is shown to out-perform conventional nearest neighbours methods on sparse case bases.

Addressing time intervals

One of the fundamental questions regarding the temporal ontology is what is time composed of. While the traditional time structure is based on a set of points, a notion that has been prevalently adopted in classical physics and mathematics, it has also been noticed that intervals have been widely adopted for expre~sion of common sense temporal knowledge, especially in the domain of artificial intelligence. However, there has been a longstanding debate on how intervals should be addressed, leading to two different approaches to the treatment of intervals. In the first, intervals are addressed as derived objects constructed from points, e.g., as sets of points, or as pairs of points. In the second, intervals are taken as primitive themselves. This article provides a critical examination of these two approaches. By means of proposing a definition of intervals in terms of points and types, we shall demonstrate that, while the two different approaches have been viewed as rivals in the literature, they are actually reducible to logically equivalent expressions under some requisite interpretations, and therefore they can also be viewed as allies.

The effect of thermal cycle profiles on solder joint damage

The relationship between the damage caused at different thermal cycles is very important. The whole of accelerated thermal cycle testing is based on the premise that damage at one cycle is representative of damage at a different cycle. In this paper, the relative damage caused by six thermal cycle profiles are predicted using Finite Element (FE) modelling and the results validated against experiments. Both creep strain and strain energy density were used as damage indicators and creep strain was found to correlate better with experiment. The validated FE model is then used to investigate the effect of altering each of the thermal profile parameters (ramp and swell times, hot and cold temperatures). The components used for testing are surface mount resistors - 1206, 0805 and 0603. The solders investigated are eutectic SnAgCu and eutectic SnAg.

The role of joint training in practitioner development for learning disability services

The presentation explores and evaluates an innovation in education and training in which two different professional trainings (nursing and social work) are integrated to produce jointly qualified specialist practitioners.