Knowledge acquisition based on rough set theory and principal component analysis

A knowledge acquisition algorithm (KA-RSPCA), based on rough set theory and principal component analysis, that increases the obtained rules' generalization power by integrating a measure of every condition attribute's condition contribution to the state space, is discussed. The process of this algorithms includes standardizing the variables, calculating the sample correlation matrix, and calculation of principal components. KA-RSPCA is apt to acquire a reduct consisting of more important contributes with a larger contribution. This algorithm is claimed to be a meaningful procedure for studying knowledge acquisition and applying it to business intelligence in various industries.