Abstract: Chest diseases are a subgroup of respiratory system diseases. The symptoms of these diseases are similar and this makes the diagnosis process difficult. Therefore, to ease the diagnosis process we gathered Information about fourteen diseases which attack the chest with their symptoms and investigations about them. In this study, we present an Arabic ontology-based approach for chest diseases diagnosis. We focus on ontology building process. This ontology can be used to help physicians and other users, determine the chest disease that a patient is suffering from and what are the investigations that should be applied. While experts can easily gather information from this data, lay users lack the expertise needed to deal with it. Most of the efforts to solve this problem focus on the English language. So also, in this study, we present a new approach based on natural language processing to translate Arabic language query to SPARQL query.