Knowledge perspectives in data grids

In this paper a methodology for accesing scientific data repositories on data grids is proposed. This methodology is based on ontology specification and knowledge representation. The concept of Knowledge Perspective is introduced, as the action of applying particular scientific conjectures or theories to the interpretation of experimental data and information. Data grid environments provide high levels of security and virtualization, which allow the users to create new data services on the data server side. These new services are based on the user’s knowledge perspective. An implementation of this concept is presented, on a Globus-enabled Java execution platform. IFIP International Conference on Artificial Intelligence in Theory and Practice - Knowledge Acquisition and Data Mining Red de Universidades con Carreras en Informática (RedUNCI)