In this paper, we present a semantic search approach based on Case-based modular Ontology. Our work aims to improve ontology-based information retrieval by the integration of the traditional information retrieval, the use of ontology and the case based reasoning (CBR). In fact, our recommender approach uses the CBR with ontology for case representation and indexing. Ontology-based similarity is used to retrieve similar cases and to provide end users with alternative recommendations. The main contribution of this work is the use of a CBR mechanism and an ontological representation for two purposes: Resource Retrieval from Web and ontology enrichment from cases.