Metatools for knowledge acquisition

Four prototype metatools, Protege, Dots, Dash, and Spark, which researchers are using to experiment with the automatic generation of knowledge-acquisition tools, are discussed. Protege and Dots are stand-alone metatools; Dash and Spark are subsystems. Dash is part of Protege II, a design environment for knowledge-based systems. Spark is part of the Spark, Burn, and Firefighter framework for the design of application systems. The two stand-alone tools, their environments, and their subsystems are compared. Protege demonstrates how one can instantiate knowledge-acquisition tools from a description of a problem-solving method. Dots, on the other hand, lets one design knowledge-acquisition tools for many applications. Spark, Burn, and Firefighter are similar to Protege II in that they emphasize developing problem-solving methods from reusable components, although Spark, Burn and Firefighter associate a knowledge-acquisition tool with each method in the library. Protege II uses Dash to generate knowledge-acquisition tools from domain ontologies.<<ETX>>