Abstract: In semantic web, the information flow obtained from different relations is certain and processing those data across the relations are not easy without proper understanding about the semantic mapping between them. It is a complex process to manually identify these mappings and it is not possible over the web. It is required to develop tools for supporting relation mapping for the success of the semantic web. A technique named sealant is designed for machine based learning for identifying the mappings. For two given catalogs, the percept in one relation is identified by sealant and it predicts the most common percepts in other catalogs. A probability based explanations for many resemblance measures are viewed using sealant which works well with all of them. Furthermore, the sealant employs different learning techniques each of which utilizes several information types either in the data occurrence or in the catalog framework of the relations. The matching precision can be enhanced by expanding the sealant for integrating sound understanding and domain restrictions into the matching process. The technique varies with its working ways using clearly explained resemblance perception and effective integration of several types of understanding. The sealant is expanded for identifying difficult mappings between the relations and explains the analysis for its effective utilization.
S. Raja Ranganathan, M. Marikkannan and S. Karthik, 2016. An Attempt for Content Based Matching on Semantic Web Using Relation Map Based Algorithmic Approaches. Asian Journal of Information Technology, 15: 840-845.