Książka = Book ; KS/2/1992/R12

83-102 stron ; 21 cm

Bibliografia s. 102

The main aim of the symbolic approach in data analysis is to extend problems, methods and algorithms used on classical data to more complex data called "symbolic objects" which are well adapted to representing knowledge and which can "unify" unlike usual observations which characterize "individual things" . We focus herc on boolean and probabilist objects and we briefly present some of their qualities and roperties. We finally develop in the context of symbolic analysis of a classical data table, a factorial axis characterisation as a probabilist object; it completes the usual vectorial representation which is not so explicit for the standard user. We particularly show the application of learning algorithms to explain multiple correspondence analysis axis which are so useful for enquiry treatments.

Bibliography p. 102

83-102 pages ; 21 cm