Automated & Semi-Automatic Knowledge Acquisition

There are three different channels applied for knowledge acquisition - for accessing and accumulating information and knowledge objects:

• manual input of objects by the administrator, assisted by semi-automatic tools
• adapting new objects provided by the business partnersvia the agent based service for updating the local and global dictionaries
• meta search engine searches for objects related to predefined context provided by the CTVs within the subject trees

Fig. 1: At least three channels for knowledge acquisition

The manual input (A) by the administrator is assisted by several tools for handling objects and configuring the subject trees (see "trainer" "clusterer"). The OTV generation and the consecutively processed subject tree referencing procedure is realized as a semi-automatic tool.

Whereby in the second channel (B) the objects provided by the update service are handled fully automatically: the OTV generation and the similarity & relationship analysis. Also in case (C): the meta search engine processes different lists generated out of the CTVs and provides information objects, which are processed automatically in the rule based filtering and classifier module (see "categorizer").