Debugging user interface descriptions of knowledge-based recommender applications

Abstract

The complexity of product assortments offered by e-Commerce platforms requires intelligent sales assistance systems alleviating the retrieval of solutions fitting to the wishes and needs of a customer. Knowledge-based recommender applications meet these requirements by allowing the calculation of personalized solutions based on an explicit representation of product, marketing and sales knowledge stored in an underlying recommender knowledge base. Unfortunately, in many cases faulty models of recommender user interfaces are defined by knowledge engineers and no automated support for debugging such process designs is available. This paper presents an approach to automated debugging of faulty process designs of knowledge-based recommenders which increases the productivity of user interface development and maintenance. The approach has been implemented for a knowledge-based recommender environment within the scope of the Koba4MS project.