Identifying and semantics of KB are playing important

Identifying Key Concepts within an Ontology Automatically identifying the key concepts within an ontology applying topological and lexical criteria including Name Simplicity, Density, Coverage, and Popularity is the main idea behind 25 approach. The ultimate goal in Pires et al. 25 model is to return a subset of concept from an ontology that match as much as possible to those concepts produced by human experts. This model is focusing more on returning key concepts within an ontology without pay that much attention to generating a graph of extracted important concepts. F. Ontology Understanding without Tears Troullinou et al. in 22 proposed an advanced version of RDF Digest: Efficient summarization of RDF/S KBs 32 as a new automatically high quality RDF/S Knowledge Bases summary producer. Finding the most representative concepts within schema graph considering the corresponding instances is the key point in generating the summary for an ontology. In this context, the structure of the graph and semantics of KB are playing important roles in the final summary. In the proposed approach, the importance of each node (concept) is determined through the Relative Cardinality, in the next step the centrality of each node in the KB is estimated by combining the Relative Cardinality with the type and number of the incoming and outgoingedges in the schema. The final step in this approach is generating valid sub-schema graphs that cover more relevant nodes by minimizing their overlaps. Two algorithms that try to optimize the local and global importance of the selected paths are applied in order to generate the final summarized sub-schema graphs. V. Conclusions In this paper, we investigate different ontology summarization measures and models. While there are automatic methods to generate ontology summaries based on different requirements and tasks but there is still a room to extract more reliable summaries. To best of our knowledge, the procedure of extracting summaries in the current methods is static which means that the summaries are produced based on some predefined measures. In an ideal case, the ontology summarization technique needs to be more flexible in the way that users or applications be able to tune the model in order to generate different summaries based on different requirements. The available approaches apply extractive technique in order to generate the final summary (the exact nodes from the original ontology are selected as a summary). non-extractive ontology summarization is a new direction in this area which can be applicable in various applications such as ontology tagging.