Modeling of products sale velocity and products recommended structure on the basis of a data reduction method

Abstract

This article presents an algorithm for the selection of the most informative product parameters, so that the product sale velocity can be predicted. The method is based on statistical information about sales with the help of data mining procedures. The constructed decision tree allows one to refer an analyzable product according to its attribute values to a cluster of sale velocities. A decision tree structure gives developments of new product information about what attributes of products influence sales most essentially. It provides direct search to the designers.