Most text search engines are intended for identification of multiple depositories storing the text of a user's query. I need one that does the opposite. Imagine we have two databases with overlapping coverage and we need to know what content is represented in only one of them. One way to do it would be by creating a file with information items extracted from one of these databases and uploading this file to a search engine which returns a list of those items of the uploaded ones that were not found in another database. An implementation of this approach exists (Batch Citation Matcher), but its application is limited to the database embedding it (PubMed). Is there an analogues application which can be applied to any website?