Advances in Applied Artificial Intelligence (Fulcher, 2006)

Copyright (c) 2013 John L. Jerz

Advances in Applied Artificial Intelligence (Fulcher, 2006)

Computational Intelligence and Its Applications

Book Description
Whether any one technology will prove to be the central one in creating artificial intelligence, or whether a combination of technologies will be necessary to create an artificial intelligence is still an open question, so many scientists are experimenting with mixtures of such techniques. In Advances in Applied Artificial Intelligence these questions are implicitly addressed by scientists tackling specific problems which require intelligence in both individual and combinations of specific artificial intelligence techniques.
Advances in Applied Artificial Intelligence includes extensive references within each chapter which an interested reader may wish to pursue. Therefore, this book can be used as a central resource from which major avenues of research may be approached.

About the Author
John Fulcher is currently a professor of information technology in the School of IT and Computer Science and director of the Health Informatics Research Centre at the University of Wollongong, Australia. He holds a BEE with Honours from the University of Queensland (1972), a Research Masters from LaTrobe University, Melbourne (1981), and a PhD from the University of Wollongong (1999). He is a member of the Association for Computing Machinery and a senior member of the Institute of Electrical and Electronic Engineers. Professor Fulcher was an invited keynote speaker at the 5th National Thai Computer Science and Engineering Conference. His research interests include microcomputer interfacing, computer science education, artificial neural networks (especially higher-order ANNs), health informatics, and parallel computing.

p.2 Several decision support systems have been developed in various fields... Usually previous experience or expert knowledge is often used to design decision support systems... A database is created from the available data and human knowledge. The learning process then builds up the decision rules. The developed rules are further fine-tuned, depending upon the quality of the solution, using a supervised learning process.

p.3 Implementation of a reliable decision support system involves two important factors: collection and analysis of prior information, and the evaluation of the solution... An object of the decision problem is also known as the decision factor.

p.5 How can human knowledge be extracted to a database? ...The human knowledge can be analysed and converted to an information table.

p.235 In this modern world, information is collected all the time: from our shopping habits to web browsing behaviours, from the calls between businesses to the medical records of individuals, data is acquired, stored, and gradually linked together. In this morass of data, there are many relationships that are not down to chance, but transforming data into information is not a trivial task. Data is obtained from observation and measurement, and has no intrinsic value. But from it we can create information: theories and relationships that describe the relationships between observations. And from information we can create knowledge: high level descriptions of what and why, explaining and understanding the fundamental data observations. The mass of data available allows us to potentially discover important relationships between things, but the sheer volume dictates that we need to use the number crunching power of computers to assist us with this process.

Data mining, or knowledge discovery as it is sometimes called, is the application of artificial intelligence and statistical analysis techniques to data in order to uncover information. Given a number of large datasets, we are fundamentally interested in finding and identifying interesting relationships between different items of data... Whatever the domain of the data, we are engaged in a search for knowledge, and are looking for interesting patterns in the data... Interest, like beauty, is in the eye of the beholder. For this reason, we cannot leave the search for knowledge to computers alone. We have to be able to guide them as to what it is we are looking for, which areas to focus their phenomenal computing power on. In order for data mining to be generically useful to us, it must therefore have some way in which we can indicate what is interesting and what is not, and for that to be dynamic and changeable.

p.236 Since "interesting" is essentially a human construct, we argue that we need a human in the data mining loop; if we are to develop an effective system, we need to allow them [humans] to understand and interact with the system effectively.

p.273 Nature is a wonderful source of inspiration for building models and techniques for solving difficult problems in design, optimisation, and control.

p.279 The fitness/objective function not only represents the quality of each solution, but acts as a link between the optimisation algorithm and the problem under consideration. It is imperative to select a good fitness function that accurately represents, in a single number, the goodness of the solution. Further, it is expected that the selected fitness function should exhibit a functional dependence that is relative to the importance of each characteristic being optimised.

Enter supporting content here