I’m evaluating Opus 12. Have one question. When searching in Windows Explorer there is a Relevance field that ranks the search hits. In Opus 12 there only seems to be the normal fields, like filename, size, etc. Is this correct? Or can I list hits based on relevance after a search in Opus 12? I don't want to manually go through 100s of files after searching. I want them ranked based on relevance. I have looked at the fields and I don't see any Relevance field.
There isn't a way to do that. I'm not sure how Explorer decides which files are more relevant either; something either matches a search query or it doesn't, in general.
I really like that I can use the Ranking (five star) on your product. If you somehow can copy the Relevance field into your product, I will buy immediately.
I have 10,000 article files. When I search, the Microsoft algoritm is pretty good. I normally get hundreds of hits and that is true for your product as well. Need a way to sort
But how is "relevance" defined here? What makes one file more relevant than another?
What kind of search are you doing? Names, contents, etc.?
@Mats: Are you using the Search field at the top-right of the window (that uses Windows Search), or the Tools > Find Files panel (that uses Opus's own find code)?
For results that come from Windows Search, we may be able to feed the Relevance column through from the shell, without having to know what it means and how it's calculated.
For our own Find Files panel, we would not be able to do that, of course.
I used Windows Search. If I use your File Files it takes foreever (>2 minutes) to get through my some 5-10,000 file. In contrast when I use Explorer search is instant. (Actually this is quite new in Windows Explorer. It used to take time there as well]
Here is a typical file name. The file consist of one academic article.
Hortacsu Syverson 2007 J Pol Econ
When I search "Syverson" it gets the same relevance as when I search "Hortacsu". This is nice because name order is random. Explorer would also list a file high if the author is not listed in the file name. I don't know how this happens. Maybe there is an author meta-tag that Explorer searches. So decent identification and instant search. All in Explorer gives ~150 hits in this case.
Your own search is not very good. It just took ~4 min to complete. It found 10 files with "Syverson" in the name - but Explorer found 18 files with "Syverson" in the filename searching the same data. But instant result.
I sometimes use Copernic and that software, I think, searches the whole file. That is also good, but I would consider that a separate issue. Explorer's Relevance field is not so detailed.
Windows Search will be faster in a lot of cases because the file contents are pre-indexed (so only the index is searched, not the files themselves). Note that if you use the search field in the top-right corner of the Opus window, this is using Windows Search behind the scenes, and so the results should be the same as in Explorer.
What are the search criteria when you're using the Find Panel? Could you show a screenshot?
If you're only searching file names, not content, it should be as fast as the disk can list the folders, which doesn't usually take very long (unless the disk is slow or over a slow network, etc.)
If contents are being searched then it will be slow unless the search is narrowed down.
Make sure the option to search inside archives is off, as that will make things very slow and usually isn't wanted.
Okay, did a mistake with Find Panel. When entering my search correctly under Name matching, the search is very quick.
For my purpose this is pretty good. The Explorer Relevance is slightly better as it also finds the original author even if I don't label the file with author name. Of course no guarantee that all such files are found.
1 Like