I have a Sharepoint foundation 2010 site with Knowledgelake software (add-ins that allow for scanning and indexing of paper documents) as well as Search Server 2010 Express.
In the past few days we have noticed that searching for some documents by metadata exposed to the search server is showing that they don’t exist, when looking in sharepoint itself you can find the document (filter by client number and it shows up) but both a search within the sharepoint website and search webservice don’t show them.
On the sharepoint website search results there is a “View Duplicates” which then shows the other documents. These documents are physical scans of different peices of paper, with hand writing on them so i can’t see why sharepoint would group them together.
If someone could inform me as to why documents would be grouped together, and or how to change the search server to ignore the duplicate grouping and show all results i would really appreciate it.
The duplicates are determined based on the calculated fcoid managed property during the search. Part of the duplicate calculation uses the DocSignature managed property. The fcoid is used to retrieve the duplicates when clicking on the “View Duplicate” links. There is no documentation on how this calculation is made. You can read more here:
http://msdn.microsoft.com/en-us/library/microsoft.office.server.search.query.keywordquery.trimduplicatesincludeid.aspx
If you want to turn off the trimming of duplicates in the SharePoint search results page by editing the Search Core Results webpart. Find the “Results Query Options” section and uncheck the the “Remove Duplicate Results” check box. Click OK and save your changes to the search results page.