Plagiarism detection is a form of detecting subject-sameness.
If you think of a document as a subject and say 95% of it is the same as another document, you could conclude that it is the same subject. (Or set your own level of duplication for subject-sameness.)
One of the early use cases for topic maps was avoiding the duplication of documentation (and billing for the same) for defense systems.
Detecting self-plagiarism from a law firm, vendor, contractor, consultant is one thing.
Putting those incidents together across a government agency, business, institution, or enterprise is a job for topic maps.