Optimization of data integration using schema matching of linguistic-based and constraint-based in the university database

University requires the integration of data from one system with other systems as needed. This is because there are still many processes to input the same data but with different information systems. The application of data integration generally has several obstacles, one of which is due to the diversity of databases used by each information system. Schema matching is one method that can be used to overcome data integration problems caused by database diversity. The schema matching method used in this research is linguistic and constraint. The results of the matching scheme are used as material for optimizing data integration at the database level. The optimization process shows a change in the number of tables and attributes in the database that is a decrease in the number of tables by 13 tables and 492 attributes. The changes were caused by some tables and attributes were omitted and normalized. This research shows that after optimization, data integration becomes better because the data was connected and used by other systems has increased by 46.67% from the previous amount. This causes the same data entry on different systems can be reduced and also data inconsistencies caused by duplication of data on different systems can be minimized.