If you use IBM® Content Classification, you can improve search quality in Watson Explorer Content Analytics by importing data from Content Classification. You can also train Content Classification knowledge bases by importing analyzed documents that you export from Watson Explorer Content Analytics.
If you configure a collection to use the provided Content Classification annotator, metadata can be added to documents when they are processed in the document processing pipeline. The metadata can improve search quality by returning highly relevant documents in the query results. The metadata can also be used as facets, which can help user narrow results to specific documents of interest.
You can export documents from Watson Explorer Content Analytics and use them to train a Content Classification knowledge base. When you export documents from an application, the system exports one file that contains multiple documents. When documents in a file exceed a certain limit, the file is divided into multiple files. The system also exports a catalog.xml file that describes the fields in the documents.
If you import the document XML files and catalog.xml file into Classification Workbench, you can use the data to train knowledge bases and decision plans. By repeatedly exporting documents over time, you can iteratively train and improve classification in IBM Content Classification.