Classification Workbench enables you to create and analyze a single-language knowledge base or a multilingual knowledge base using any combination of supported languages.
See What languages are available? for a list of supported languages.
A multilingual knowledge base requires a hierarchical knowledge base structure. You can manually create the hierarchical structure using the embedded Knowledge Base Editor application (see Knowledge Base Editor). Alternatively, the system can create the hierarchical structure for you (see Multilingual knowledge bases). If you would like to use a different hierarchical knowledge base structure (for example, from a different project), use the Import Structure option on the Knowledge Base Editor File menu.
You can specify the language by clicking the Update button and selecting one or more languages in the Project Options window (see Global options).
Classification Workbench identifies the language of each content item, to ensure proper analysis. More specifically, the system determines the language of text in fields for which the content type was specified (see More about natural language processing). If a field contains text in more than one supported language, the predominant language (that is, the language in which most of the text is written) is identified for the item, and analysis is based on this language.