Mohammad Javad Kargar - Department of Computer Engineering, Faculty of Engineering, University of Science and Culture, Tehran
Samira Babalou - Department of Computer Engineering, Faculty of Engineering, University of Science and Culture, Tehran
Seyyed Hashem Davarpanah - Department of Computer Engineering, Faculty of Engineering, University of Science and Culture, Tehran
Pazir Sarafraz - Faculty of Computer Science and Engineering, Shahid Beheshti University, Tehran, Iran,

The increasing popularity and extension of semantic web applications have led to myriad amounts of RDF data and ontologies. The large-scale ontology and complex RDF datasets are associated by several sorts of complexities. It is so difficult for users to understand these data sets even if using visualization tools. In order to promote the process and make large-scale ontologies more understandable, ranking algorithms have been used. To this end, in this paper, we introduce a Neural Network-based ranking approach which exploits centrality measures, number of children, and hierarchy level among ontology concepts. The evaluation shows higher performance compared to existing methods.

Ontology, Rank, Query answering, Neural network, Ontology summarization