This paper provides an overview of a research and development project on the automated testing of complex telecommunication networks. The paper reports on the current state of the project, which, for example, includes the development of a database for the organisation and the management of the data that is generated in a network test. Along this line the paper also briefly reports on a future direction in the project, a data warehouse for the management and the organisation of data that is gathered in the wider scope of the project. A major focus of the paper is the proposal and description of a decision support system (DSS). The overall aim of the DSS is to support the decision making of test engineers to further improve the automated testing of complex telecommunication networks.

Learning probabilistic networks for expert system applications has seen a great surge of research activity in recent years. The paper reports on the adoption of probabilistic networks for verifying the pass/fail result of automated testing at the software verification stage in the development lifecycle of high-speed telecommunications equipment. The focus is on learning Bayesian belief networks (BBNs) from automated test data on a per test execution basis. This facilitates result classification and assurance, taking advantage of their graphical nature to provide accountability for the decision.

This paper compares three different data analysis methods in a subfield of the coronary heart disease risk assessment (CHDRA) area—the identification of increased blood cholesterol levels. A data set containing the cholesterol data of 166 persons is employed as a test case, and analyzed in three experimental investigations. The first analysis method employs a predominantly knowledge-based fuzzy expert system solution to the problem. The second method employs statistical discriminant analysis on the same data, whereas the results of the third analysis are obtained by an artificial neural network. This study evaluates the advantages as well as the disadvantages of each method. Special attention is given to the systems' individual capability for managing the uncertainty involved in the decision-making process. The results achieved in the study provide evidence for the complementary and mutually supporting character of the different approaches. ©2000 John Wiley & Sons, Inc.

Both information retrieval and case-based reasoning systems rely on effective and efficient selection of relevant data. Typically, relevance in such systems is approximated by similarity or indexing models. However, the definition of what makes data items similar or how they should be indexed is often nontrivial and time-consuming. Based on growing cell structure artificial neural networks, this paper presents a method that automatically constructs a case retrieval model from existing data. Within the case-based reasoning (CBR) framework, the method is evaluated for two medical prognosis tasks, namely, colorectal cancer survival and coronary heart disease risk prognosis. The results of the experiments suggest that the proposed method is effective and robust. To gain a deeper insight and understanding of the underlying mechanisms of the proposed model, a detailed empirical analysis of the models structural and behavioral properties is also provided.

This paper presents a method of partitioning a Markov tree of belief functions into clusters so as to efficiently implement parallel belief function propagations on the basis of the local computation technique. Our method initially represents computations of combining evidence on all nodes in a Markov tree as parallelism instances, then balances the computation load among these instances, and finally partitions them into clusters which can be mapped onto a set of processors in a PowerPC network. The advantage of our method is that the maximum parallelization can still be achieved, even with limited processor availability.

Two graphical subsystems that make use of the novel concurrent architecture of the INMOS transputer series including the G170 Colour Pallette are presented. To achieve maximum performance and flexibility from the hardware, the subsystems are programmed in occam, a concurrent language developed in close association with the transputer. The subsystems at present interface to concurrent simulations of the human basilar membrane and vocal tract, implemented in occam2 on multitransputer arrays. Particular emphasis is placed on the application of concurrency to the physiological processing of speech and image presentation for medical research. A key aspect of the approach is the scope for both software and hardware modularity. With respect to hardware this means that an architecture can be tailored to meet cost/performance requirements. A preferred arrangement of transputers is presented.>