Artificial intelligence techniques for rational decision making (Q483306)

The title of the book is very challenging and covers a broad field; many practical applications can be based on these concepts. And quite naturally it is impossible to discuss all these problems in one book. These difficulties are solved by the author in a very readable way; he breaks the field down into small parts, each part including some mathematical/theoretical background and application areas. The theoretical background as well as the respective applications are presented in a relatively short way, but they can successively be used as examples for other applications. This can be seen by the introduction itself which indicates the application of neural networks, fuzzy sets, rough sets, Hopfield networks, support vector machines, genetic algorithms, particle swarm optimization, and simulated annealing. In short presentations it is explained: how to build causal functions, how to use correlation functions, how to deal with missing data, rational counterfactuals, how to handle flexibly-bounded rationality, the filtering of irrelevant information and group decision making. The examples have been taken from very interesting areas, e.g., (militarized) interstate disputes, epileptic activity, antenatal data, among others. Each chapter has its own very detailed list of references. This book shows in a very impressive way many applications of artificial intelligence methods and should be studied very carefully. But it shows also the big dilemmas for real-world applications: which model should be used in a given situation, how to get the required mathematical knowledge, how to create the interdisciplinary cooperation between these two areas, the software implementation for the given problem.