Search of terms for integration of semantically similar data

The paper has considered issues of semantically similar data base integration in the framework of a new information system. The issues are connected with determination of attributes having similar domains. The algorithm for search of such attributes has been suggested. The algorithm, unlike most modern algorithms, takes into account not only the syntactic similarity of attribute names, but also the semantic similarity. The degree of similarity is determined by the number of identical attribute values. The similarity found allows you to combine databases created for one domain at different times on different DBMSs and covering different but interrelated parts of the data domain. Valuable information accumulated over a long period in old databases will be used correctly in the new database.