We have added full example how to autogging documents based in metadata database values. http://wiki.openkm.com/index.php/Simple_autotagging
This is the most easiest way for autotagging documents. Take it as is, a simple aproximation to autotagging feature. In this case we used simply comparasion looking for text into documents. Obviously are more complex ways for doing it, based in natural languages, synonims etc... for that should connect with thirdparty applications to analyze content and extract key values or implement your own. The example gives a good startup to take in mind some ideas how implement your own autotagging extractor engine.
This is the most easiest way for autotagging documents. Take it as is, a simple aproximation to autotagging feature. In this case we used simply comparasion looking for text into documents. Obviously are more complex ways for doing it, based in natural languages, synonims etc... for that should connect with thirdparty applications to analyze content and extract key values or implement your own. The example gives a good startup to take in mind some ideas how implement your own autotagging extractor engine.