SemMed: Applying Semantic Web to Medical Recommendation Systems Alejandro Rodríguez¹, Enrique Jiménez¹, Jesús Fernández¹, Martin Eccius¹, Juan Miguel Gómez¹, Giner Alor-Hernandez², Rubén Posada-Gomez², Josef Noll 3 , Carlos Laufer 4 ¹Departamento de Informática, Universidad Carlos III de Madrid {alejandro.rodriguez, enrique.jimenez, jesus.fernandez.prieto, martin.eccius, juanmiguel.gomez}@uc3m.es ²Division of Research and Postgraduate Studies, Instituto Tecnológico de Orizaba, Mexico {galor, rposada}@itorizaba.edu.mx 3 University Graduate Center (UniK), Universidad de Oslo josef@unik.no 4 Pontificia Universidade Católica, Rio de Janeiro laufer@globo.com Abstract The systematic method used to identify diseases is called differential diagnosis (DDx). It is mostly used by healthcare professionals to diagnose a specific disease in a patient. If the disease is diagnosed, the method could be used to recommend the medications in order to receive treatment. The goal of the current paper is to design a system based on Semantic Web Technologies to develop a system with the capability to assist healthcare professionals regarding the possible medication or drug to prescribe, according to following fundamental selection criteria. Keywords: drug, semantic web, recommendation system, medicine 1. Introduction Currently, numerous Expert and Decision Support Systems (DSS) allow medical collaboration, like in Differential Diagnosis (specific or general). In the semantic web field this aspect has not yet been extensively developed, and although there are some publications on generation of medical ontologies [1], this aspect is still rather analyzed in previous work. A medical recommendation system based on SW technologies has not yet been developed which initiated this research paper. However, it should be mentioned that there are several system references about medicine or active ingredient interactions, but their final goal is not the drug recommendation ([2], [3], [4]). This paper is structured as follows. Section 2 describes the system in detail, explaining how it functions and all of the components which are involved in the system such as variables, the working of the system and use case. Section 3 explains the system architecture, which is composed of 4 subsystems: the Ontology Manager subsystem, which will have all the necessary information to perform inference in order to obtain the drugs; the Rules Manager which works together with the Ontology Manager and Inference Engine to provide knowledge rules; the Inference Engine, which is the kernel of the system because it allows the combining of the ontology manager and knowledge rules data in order to infer the results, and finally a support DB which contains additional information about drugs, allergies, etc, which complement ontology information. In section 4 is showed a graphic about how ontology is structured and the relationships that the ontology may contain. In the section 5 we talk about the rules that the system needs to infer correct results are shown with the conjunction of the ontology and inference engine and an example of the rule structure. Section 6 discusses related work in this topic and section 7 provides conclusions and future work of the system. 2. Solution Prescribing drugs may initially appear a trivial task; however it is a complicated process. Multiple factors