An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

From the Publisher:
"Covering the latest issues and achievements, this well documented, precisely presented text is timely and suitable for graduate and upper undergraduate students in knowledge engineering, intelligent systems, AI, neural networks, fuzzy systems, and related areas. The author's goal is to explain the principles of neural networks and fuzzy systems and to demonstrate how they can be applied to building knowledge-based systems for problem solving. Especially useful are the comparisons between different techniques (AI rule-based methods, fuzzy methods, connectionist methods, hybrid systems) used to solve the same or similar problems."
-- Anca Ralescu, Associate Professor of Computer Science, University of Cincinnati Neural networks and fuzzy systems are different approaches to introducing human-like reasoning into expert systems. This text is the first to combine the study of these two subjects, their basics and their use, along with symbolic AI methods to build comprehensive artificial intelligence systems. 
In a clear and accessible style, Kasabov describes rule- based and connectionist techniques and then their combinations, with fuzzy logic included, showing the application of the different techniques to a set of simple prototype problems, which makes comparisons possible. A particularly strong feature of the text is that it is filled with applications in engineering, business, and finance. AI problems that cover most of the application-oriented research in the field (pattern recognition, speech and image processing, classification, planning, optimization, prediction, control, decision making, and game simulations) are discussed and illustrated with concrete examples.
Intended both as a text for advanced undergraduate and postgraduate students as well as a reference for researchers in the field of knowledge engineering, Foundations of Neural Networks, Fuzzy Systems, and Knowledge Engineering has chapters structured for various levels of teaching and includes original work by the author along with the classic material. 
Data sets for the examples in the book as well as an integrated software environment that can be used to solve the problems and do the exercises at the end of each chapter are available free through anonymous ftp.

Foundations of neural networks, fuzzy systems, and knowledge engineering

This paper introduces evolving fuzzy neural networks (EFuNNs) as a means for the implementation of the evolving connectionist systems (ECOS) paradigm that is aimed at building online, adaptive intelligent systems that have both their structure and functionality evolving in time. EFuNNs evolve their structure and parameter values through incremental, hybrid supervised/unsupervised, online learning. They can accommodate new input data, including new features, new classes, etc., through local element tuning. New connections and new neurons are created during the operation of the system. EFuNNs can learn spatial-temporal sequences in an adaptive way through one pass learning and automatically adapt their parameter values as they operate. Fuzzy or crisp rules can be inserted and extracted at any time of the EFuNN operation. The characteristics of EFuNNs are illustrated on several case study data sets for time series prediction and spoken word classification. Their performance is compared with traditional connectionist methods and systems. The applicability of EFuNNs as general purpose online learning machines, what concerns systems that learn from large databases, life-long learning systems, and online adaptive systems in different areas of engineering are discussed.

Evolving fuzzy neural networks for supervised/unsupervised online knowledge-based learning

We describe an investigation into e-mail content mining for author identification, or authorship attribution, for the purpose of forensic investigation. We focus our discussion on the ability to discriminate between authors for the case of both aggregated e-mail topics as well as across different e-mail topics. An extended set of e-mail document features including structural characteristics and linguistic patterns were derived and, together with a Support Vector Machine learning algorithm, were used for mining the e-mail content. Experiments using a number of e-mail documents generated by different authors on a set of topics gave promising results for both aggregated and multi-topic author categorisation.

Mining e-mail content for author identification forensics

The Discovery of Spoken Language, Peter W. Jusczyk (Ed.). MIT Press (1997), ISBN 0 262 10058 4

Software source code authorship analysis has become an important area in recent years with promising applications in both the legal sector (such as proof of ownership and software forensics) and the education sector (such as plagiarism detection and assessing style) Authorship analysis encompasses the sub-areas of author discrimination, author characterization, and similarity detection (also referred to as plagiarism detection) While a large number of metrics have been proposed for this task, many borrowed or adapted from the area of computational linguistics, there is a difficulty with capturing certain types of information in terms of quantitative measurement Here it is proposed that existing numerical metrics should be supplemented with fuzzy-logic linguistic variables to capture more subjective elements of authorship, such as the degree to which comments match the actual source code’s behavior These variables avoid the need for complex and subjective rules, replacing these with an expert’s judgement Fuzzy-logic models may also help to overcome problems with small data sets for calibrating such models Using authorship discrimination as a test case, the utility of objective and fuzzy measures, singularly and in combination, is assessed as well as the consistency of the measures between counters

A fuzzy logic approach to computer software source code authorship analysis

Software forensics is the use of authorship analysis techniques to analyse computer programs for a legal or official purpose This generally consists of plagiarism detection and malicious code analysis IDENTIFIED is a system that has been designed to assist with the extraction of count-based metrics from source code, and with the development of models of authorship using statistical and machine-learning approaches Software forensic models can be used for identification, classification, characterisation, and intent analysis One of the more promising methods for identification is case-based reasoning, where samples of code can be compared to those collected from known authors 1 SOFTWARE FORENSICS The frequency and severity of the many forms of computer-based attacks such as viruses and worms, logic bombs, trojan horses, computer fraud, and plagiarism of software code (both object and source) have all become increasingly prevalent and costly for many organisations and individuals involved with information systems Part of the difficulty experienced in collecting evidence regarding the attack or theft in such situations has been the definition of appropriate measurements to use in models of authorship and the development of appropriate models from these metrics Several options for developing such models for identification, discrimination, characterisation, and intent analysis exist, including subjective expert opinion (including fuzzy logic as in [Kilgour, et al, 1997]) and statistical and machine learning models using formally defined metrics Each offers its own set of advantages and disadvantages for the task of software forensics With the difficulties of data collection and the goal of increasing the accessibility of such modelling techniques in mind a system called IDENTIFIED is being developed It is intended to assist with the task of software forensics which is defined here to be the use of software code authorship analysis for legal or official purposes [Gray, et al, 1997] IDENTIFIED uses combinations of wildcards and special characters to define count-based metrics, allows for hierarchical meta-metric definitions, automates much of the file handling task, extracts metric values from source code, and assists with the analysis and modelling process In particular IDENTIFIED will enable the analyst to use several analysis procedures such as case-based reasoning It is hoped that the availability of such tools will encourage more detailed research into this area of ever-increasing importance 2 SOFTWARE FORENSICS Source code is the textual form of a computer program that is written by a computer programmer in a computer programming language These programming languages can in some respects be treated as a form of language from a linguistic perspective, or more precisely as a series of languages of particular types, but within some common family In the same manner that written text can be analysed for evidence of authorship (such as [Sallis, 1994]), computer programs can also be examined from a forensics or linguistics viewpoint [Sallis, et al 1996] for information regarding the program’s authorship Figure 1 [from Gray et al 1997] shows two small code fragments that were written in C++ by two separate programmers Both programs provide the same functionality (calculating the mathematical function factorial(n), normally written as n!) from the users’ perspective That is to say, the same inputs will generate the same outputs for each of these programs As should be apparent, each programmer has solved the same problem, that of calculating the factorial of an input, in both a different manner (algorithm) and with a different style exhibited in his or her code These stylistic differences include the use of comments, variable names, use of white space, indentation, and the levels of readability in each function These fragments are obviously far too short to make any substantial claims about the feasibility of using source code characteristics to make statements regarding the author(s) However, they do illustrate the fact that programmers writing programs will often do so in a significantly different manner to another programmer, without any instruction to do so Both of these functions were written in the natural styles of their respective authors // Factorial takes an integer as an input and returns // the factorial of the input // This routine does not deal with negative values! int Factorial (int Input)

IDENTIFIED: software authorship analysis with case-based reasoning

A methodology for speech data analysis and a framework for adaptive speech recognition using fuzzy n

The paper presents a general engineering model of a hybrid system for speech recognition and information retrieval. The speech recognition part consists of a low level neural network module for phoneme recognition and a higher level fuzzy reasoning module for word recognition and language modelling. There are several features such systems are characterised by, namely: adaptability, i.e. the system is adaptable to new speakers and accents; dealing with ambiguity during the recognition process; extendability, i.e. new modules can be introduced to the system at a later stage and the dictionary is extendable. A case study of building an experimental English to Maori talking dictionary for a small dictionary of words and speaker independent mode is reported.

Richard Kilgour

Papers

A fuzzy logic approach to computer software source code authorship analysis

IDENTIFIED: software authorship analysis with case-based reasoning

A methodology for speech data analysis and a framework for adaptive speech recognition using fuzzy n

Intelligent human computer interfaces and the case study of building English-to-Maori talking dictionary

From hybrid adjustable neuro-fuzzy systems to adaptive connectionist-based systems for phoneme and word recognition