Combining Ontology Alignment Metrics Using the Data Mining Techniques

Several metrics have been proposed for recog- nition of relationships between elements of two Ontologies. Many of these methods select a number of such metrics and combine them to extract existing mappings. In this article, we present a method for selection of more eective metrics - based on data mining techniques. Furthermore, by having a set of metrics, we suggest a data-mining-like means for com- bining them into a better ontology alignment. (symmetry) For example one popular metric is Edit Distance (7) mea- suring String Similarity between entities under consideration. Also there exist metrics named Resnik Similarity (16) and Upward Cotopic distance (8) that measures Linguistic and Structural similarities and distances, correspondingly. Most of the Ontology Alignment methods select or define some met- rics and combine them to recognize existing mappings. In the extraction phase, in most of these methods, couples having Compound Similarity higher than a predefined threshold - after applying a number of constraints - are selected as final mappings. A number of such methods are explained in the next section. (4) contains a more complete list. In this article a method based on data mining techniques is proposed to select more appropriate metrics among an ini- tial set and then calculate amount of Compound Similarity