Semantic search using modular ontology learning and case-based reasoning

In this paper, we present a semantic search approach based on Case-based reasoning and modular Ontology learning. A case is defined by a set of similar queries associated with its relevant results. The case base is used for ontology learning and for contextualizing the search process. Modular ontologies are designed to be used for case representation and indexing. Our work aims at improving ontology-based information retrieval by the integration of the traditional information retrieval process, the use of ontology learning (OL) and the Case-Based Reasoning (CBR) process. In fact, the proposed approach uses the CBR with semantic Web language markup -by ontology- for case representation and indexing. Ontology-based similarity is used to retrieve similar cases and to provide end users with alternative documents 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 learning and enrichment from cases. This approach builds a knowledge corpus -- represented by ontology modules - resulting from the collaboration actions of users. The experiment shows an improvement in terms of results' precision and ontology learning relevance.