In our Customer Center publication site (SP 2010), we have uploaded two different documents that have big parts of text in common. They are similar, but not equal. Although different documents, in the search results they are returned as duplicates: only one of both documents is shown whereas the other can only be accessed by clicking on 'View Duplicates'.
After reading on some Internet, I found out that these documents are considered as duplicates due to the fact that they are considered as 'near-duplicates'.
Apparently, when documents have a certain amount of chunks of matching text or content, these matches are enough for SharePoint to consider the documents as duplicates, even if the file name or metadata of both documents are different!
This is the URL of the site where I found this information: http://www.aquaforest.com/wp/index.php/finding-duplicate-files-in-sharepoint/
Does anyone know of a method or some kind of workaround to avoid these 'false' duplicates in the search results?