In financial benchmarking, the first step is financial statement analysis to help determine which company characteristics to measure and which yardsticks to apply. However, for the task of running computerized benchmarking systems the amount of financial information required is often so large as to render comparison between companies difficult ¾ or at least very time consuming. The overall objective of this study is to investigate the potential of neural networks for pre-processing the vast amount of financial data available on companies, and for presenting the approximated financial performance position of one company as compared to that of others. The study demonstrates how a large annual reports database on international pulp and paper companies can be pre-processed, i.e. classified with self-organizing maps that is one form of neural networks. The test results are encouraging, and show that self-organizing maps are a viable tool for organizing large databases into clusters of compa...

Abstract not found

Very many things in our business and auditing environment are changing at an increasing rate. One central theme in auditing is how information technology developments affect the nature of the audit process and the audit skills. Auditors have to ask how to operate in new environments. New information technology support systems for monitoring and controlling operations could be useful. Artificial neural network (ANN) based information systems are proposed as one possible solution as a support tool for auditors. This article introduces the ANN technology and reviews the literature on auditing ANN applications. The review showed that the main application areas in auditing were material errors, management fraud, and support for going concern decision. ANNs have also been applied to internal control risk assessment, audit fee, and financial distress problems. In addition the paper summarises modeling issues of the ANN applications pertaining to auditing problems. Finally, the paper outlines possible tasks were ANN based support systems could be used within auditing.

An artificial neural network model for the recognition of Escherichia coli O157:H7 restriction patterns was designed. In the training phase, images of two classes of E. coli isolates (O157:H7 and non-O157:H7) were digitized and transmitted to the neural network. The system was then tested for recognition of images not included in the training set. Promising results were achieved with the designed network configuration, providing a basis for further study. This application of a new generation of computational technology serves as an example of its usefulness in microbiology. In recent years molecular typing methods have revolutionized the epidemiologic investigation of food-borne illnesses (17) by facilitating very specific identifications of suspected agents. The pulsed-field gel electrophoresis (PFGE) procedure (23), modified by contour-clamped homogeneous electric field technology (4), separates up to megabase-range fragments of genomic DNA after digestion with restriction enzymes. Restriction profiles of Klebsiella pneumoniae DNA, generated by PFGE, allowed the precise characterization of strains and isolates

An important problem in evolutionary computing is the design of genetic representations of neural networks that permit optimization of topology and learning characteristics. One promising approach for genetic representation of neural networks is the use of grammars to depict a process in which neural networks may be generated. Existing grammar representations of neural networks describe classes of networks with homogenous processing elements, simple fixed learning mechanisms and little organized topological structure. In previous research we have presented an attribute grammar representation for classes of networks with modular topology. Each parse tree generated by the grammar encodes a neural network specification which is subsequently executed by an interpreter. By expanding the grammar to include the control of the sequence of activity in the networks, we have been able to reduce the interpreter to a simple model of the operation of individual neurons in the networks. The expanded ...

This report is a review of the main neuro-control technologies. Two main kinds of neuro-control approaches are distinguished. One entails developing a single controller from a neural network and the other one embeds a number of controllers inside a neural network. The single neuro-control approaches are mainly system inverse: the inverse of the system dynamics is used to control the system in an open loop manner. The Multi-Layer Perceptron (MLP) is widely used for this purpose although there is no guarantee that it can succeed in learning to control the plant and that, more importantly, the unclear representation it achieves prohibits the analysis of its learned control properties. These problems and the fact that open loop control is not suitable for many systems highly restricts the usefulness of the MLP for control purposes. However, the non-linear modelling capability of the MLP could be exploited to enhance model based predictive control approaches since essentially, an accurate m...

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : viii I. Introduction : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 1-1 II. V&V Background : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 2-1 2.1 Validation vs. Verification : : : : : : : : : : : : : : : : : : : : : : : : : 2-1 2.2 Rule-Based V&V Approaches : : : : : : : : : : : : : : : : : : : : : : : 2-2 2.3 Neural Networks V&V : : : : : : : : : : : : : : : : : : : : : : : : : : : 2-4 2.4 Validation Issues : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 2-5 III. Knowledge Representation : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3-1 3.1 Uncertainty : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3-1 3.2 Bayesian Networks : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3-1 3.3 Bayesian Knowledge Base : : : : : : : : : : : : : : : : : : : : : : : : : 3-4 3.4 Knowledge Acquisition and ...

There is evidence that over a series of cortical processing stages, the visual system of primates produces a representation of objects which shows invariance with respect to, for example, translation, size, and view, as shown by recordings from single neurons in the temporal lobe (Rolls, 1992; Desimone, 1991; Tanaka et al., 1991). To clarify how such a system might learn to recognise `naturally' transformed objects, I investigate a model of cortical visual processing which incorporates a number of features of the primate visual system. The model consists of a series of layers with convergence from a limited region of the preceding layer, and mutual inhibition over a short range within a layer. The feed-forward connections provide the inputs to competitive networks, each utilising a modified Hebb-like learning rule which incorporates a temporal trace of the preceding neuronal activity. The modified Hebb-rule, called simply the trace learning rule, is aimed at enabling neurons to learn t...

A finite automaton --- the so-called neuromaton, realized by a finite discrete recurrent neural network, working in parallel computation mode, is considered. Both the size of neuromata (i.e., the number of neurons) and their descriptional complexity (i.e., the number of bits in the neuromaton representation) are studied. It is proved that a constant time delay of the neuromaton output does not play a role within a polynomial descriptional complexity. It is shown that any regular language given by a regular expression of length n is recognized by a neuromaton with \Theta(n) neurons. Further, it is proved that this network size is, in the worst case, optimal. On the other hand, generally there is not an equivalent polynomial length regular expression for a given neuromaton. Then, two specialized constructions of neural acceptors of the optimal descriptional complexity \Theta(n) for a single n--bit string recognition are described. They both require O(n 1 2 ) neurons and either O(n) con...

I thank all the individuals involved in the nomination and selection process of IAPR for this honor of the 1992 King-Sun Fu award. In addition, I thank my many collaborators from all over the world over the last three decades, as this award also recognizes their contributions. If memory serves me correctly it was in the early summer of 1961 a little over 31 years ago that I first met King-Sun Fu. King-Sun had joined Purdue as an Assistant Professor in 1960 the year I received my Ph.D. and joined General Dynamics / Electronics (GD/E). Thanks to the excitement generated by Frank Rosenblatt's