Abstract
The paper presents a comparison study of the rough sets approach and probabilistic techniques, in particular, discriminant analysis and probabilistic inductive learning, to data analysis on a common set of medical data. This study completes the comparison done in [9], by taking into account, in addition to the location model of discriminant analysis, the linear Fisherian discrimination and Bayesian tree classifiers derived via inductive learning approach. A general discussion on similarities and differences among compared methods is given. Particular attention is paid to data reduction and creation of decision rules. The outcomes of a computational experiment on the common set of data are described and discussed.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call