International Journal of Information Studies Siham Amrouch, Sihem Mostefai 4 1 2012 https://www.dline.info/ijis/fulltext/v4n1/ijisv4n1_5.pdf In the last decade, ontologies became very relevant tools for knowledge management and engineering because of their potential power to embed semantic modelling. They explicitly specify the concepts of a domain and their semantic relationships. However, ontologies are designed and developed by several designers and developers according to their special needs and requirements. Hence, we may always find that the same domain or even related domains are modelled by different ontologies. We may also find that some ontologies model overlapping domains, resulting sometimes in redundancies and/or inconsistencies between them. So, to create a common knowledge base and to avoid overlapping between existing ontologies, we have opted for ontology merging. The contribution presented in this paper provides a new algorithm for fully automated ontology merging that provides complete, consistent and coherent global merged ontology. This automation is handled by the semantic integration during the similarity identification stage. This characteristic features the main difference of our contribution with regards to most of the existing algorithms for ontology merging, such as CHIMAERA, PROMPT, ONION, FCA-Merge, GLUE, etc. Our algorithm combines lexical and semantic measures for identifying similar concepts that have to be merged into a single one in the resulting ontology. In this stage, it combines two different sub-modules, the first one analyses the concepts’ names, the second one their properties. If the first sub-module fails to find mappings the second one will accomplish the task. Hence the performances of the algorithm will be better.