Knowledge Based Service Systems

The general concept of a Service System is now used to refer to those complex systems composed of heterogeneous entities, including people, technology and business operations, which interact through the exchange of services. The aim of this paper is to introduce the concept of Knowledge Based Service Systems and demonstrate that the way to make an economic, managerial, organizational, industrial, computer system smarter, which means that the system evolves by adapting to the needs and changing conditions thereby reducing the mismatch and loss of resources, is to consider such a system as a Service System based on the management and manipulation of Knowledge. To achieve a dynamic and adaptive system, we need a description of the system, entities and services, as well as a definition of the structured knowledge based on ontologies, metadata, taxonomies and thesauri, and so a kind of Knowledge Representation which contains both the raw data, obtained by recording the behavior of the system, and the knowledge gained through the study of the domain. Consequently, we need to design a Data Mining and Knowledge, Inference and Learning system, that is capable of working on both the raw and structured data and so bringing both understanding and prediction, with the aim of making the system able to adaptively evolve thereby reducing the mismatch in the services exchange mechanism. It will be fundamental to exploit results obtained by the scientific community in the field of probabilistic methods for Collaborative Filtering which are based on techniques suitable for handling data belonging to the lowest level of the knowledge hierarchy (memory based), at the intermediate level (content based) and at the highest level (hybrid)