Affiliations: [a] Department of Mathematics, Nanjing University, Nanjing, China
| [b] Department of Mathematics, Comsats Institute of Information Technology, Attock Campus, Pakistan
Correspondence:
[*]
Corresponding author. Yaojun Chen, Department of Mathematics, Nanjing University, Nanjing 210093, China. E-mail: [email protected].
Abstract: The core problem of Ontology mapping and various kinds of ontology engineering applications is the calculation of similarity between concepts in ontology. From the machine learning point of view, by means of learning the sample set, it gets the optimal ontology similarity calculation function, so that each pair of concepts mapped to a positive real number, thus reflected the similarities between concepts. After representing the ontology using graph, the goal of ontology learning is to obtain a real-valued function, which maps each pair of vertices into real axes and uses distances to reflect the similarities between concepts of vertices. In this paper, we present an ontology learning algorithm in view of ontology geometry distance computation and deep learning tricks. The iteration procedure is designed and the experiments show the effectiveness of given ontology algorithm.
Keywords: Ontology, similarity measuring, ontology mapping, distance calculation, deep learning
