The concept behind AutoSort .


The fundamental concept:

AutoSort relies on the fact that most file names contain useful information such as a celebrity name, either in full or in part.

AutoSort has a celebrity name database containing information about thousands of celebrities.  This allows it to run the name of every file through advanced identification algorithms to see if any match can be made with it's celebrity database.  Either in full or partial matches are supported.

After the identification phase, the user is presented with the results and allowed to modify or select identifications made by AutoSort.  Should AutoSort fail to suggest the appropriate celebrity, the use can manually identify the file with a nifty little 'Identify As...' tool.

Once you have reviewed the identivications, you just tell the program to move the files in their respective folders, within your main collection folder.

Once AutoSort is done tossing the files, they will end up stored in a hierarchy of folders bearing celebrity names and organized according to your preferences.

 

How good is it?

One might think that such program would be limited to identifying file names like "ROM_CindyCrawford_003.jpg" where the celebrity name is obvious. AutoSort is much smarter than that!

File names containing a partial celebrity name such as these would also get suggested as being 'Cindy Crawford':

"ccrawford", "crawfordc", "c_crawford", "crawford_c",

"cindyc", "ccindy", "cindy_c", "c_cindy",

"crawc", "ccraw", "craw_c", "c_craw",

"cincra", "cracin", "cin_cra", "cra_cin" ,

"cinc", "ccin",

And since Cindy Crawford is so very popular, she'd also get suggested for file names containing:

 "cic", "crc", "cc", etc…

Note that several more combinations would work as well, but I think you get the picture! J

You should know however, that AutoSort does not actually 'look' at the images, so there's no way it will ever identify a file name such as "0625.jpg".  In this case, you would use the Identify as... feature.