Semaphore Text Mining Side Panel

The Semaphore Text Mining Side Panel is an innovation release of a tool that should be useful to facilitate model enhancement in the context of the document corpus over which classification is to be run.

The Semaphore Text Mining Side Panel relies on a MarkLogic database and an instance of Semaphore Classification and Language server to support its functionality. Because only one ML database is used (albeit dedicated to the Semaphore Text Mining Side Panel service), documents within a particular Semaphore Text Mining Side Panel instance will be available to all users.

It allows users to classify multiple documents from their corpus and run analyses over them. For instance it is possible to see which noun phrases co-occcur with particular classifictions so that the models can be enhanced to improve classification quality.