Probalistic Conditional Independence Structures

Citation

Hutton, D.M. (2006), "Probalistic Conditional Independence Structures", Kybernetes, Vol. 35 No. 5, pp. 764-765. https://doi.org/10.1108/03684920610662548

This is a topic that rests somewhere between statistics and artificial intelligence, which makes it of interest to both systemists and cyberneticians. It confirms the need to accept that transdisciplinary studies are essential for endeavours in all fields and that no field can remain an isolated entity. This book is by Dr Milan Studeny of the Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic. It is in the series Information Science and Statistics (ISS), whose editors combine expertise in computer science, statistics and biological cybernetics.

Firstly one should be clear about the topic “Conditional independence”, and “Conditional Independence Structures”. Here the author helps the reader by explaining in his introduction that the central topic of his book is to “describe the structures of probalistic conditional independence” in such a way that the corresponding mathematical model has both relevant interpretation and offers the possibility of computer implementation. It is, as the author writes, a mathematical monograph which found its motivation in artificial intelligence and statistics. This suggests a special readership where concepts from several branches of mathematics need to be understood and with an obvious link to measure theory, discrete mathematics, information theory and algebra. Other areas occasionally used are, for example: probability theory, mathematical statistics, topology and mathematical logic. An initial report on the book was included in Kybernetes 34, No. 7/8, 2005, pp. 1293, when it was first published. It has now been used by many of this journal's readers who believe that it does fall within the area of interest of researchers and others involved with informatics or theoretical cybernetics.