How do you ensure sufficient accuracy/relevance when generating/deriving meta data from content?

Wednesday, 02 February 2011 13:05

Semaphore comes with a Classification Review tool that allows users to test the performance of classification against a known set of content. The tool helps identify improvements to the classification strategy. Semaphore use a rules based approach to classification – which is known to be the most accurate approach. It is also very transparent and controllable – so you can see what is going on and make adjustments. There are also ways to create feedback loops with end users based on their search experience.