software and services > Subprojects > CHECKERS
Database organizer Naturalis and RCE have cooperated in CATCH projects MITCH and RICH. In MITCH, a programme has been developed that purges databases by using the regular combined appearance of values in database tables. RICH developed software that recognizes certain entities fully automatic and extracts them from free text. These named entities can then be added to existing databases, or serve as an index for free text information retrieval. In CATCHPlus, a new programme called Entity Checker will be developed. EntityChecker enables storage of relevant information (who, what, where, when) from unstructured text in a database. Besides this a module, Value Checker, is dveloped, that enables the semi-automatic detection and correction of wrong values in database fields, the completion of empty fields by using referencing structures (vocabularies), and automatic linking of unstructured text to thesauruses (annotation). This programme will be used by RCE to create durable access to the continuous flow of archaeological reports. The programme can also be used by other institutions or persons with a database or collection of unstructured text.
