NetWeaver Help

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
NetWeaver is a knowledge base development system developed for the Microsoft Windows platforms that provides a graphical environment in which to construct and evaluate knowledge bases. 

The software provides  graphic tools for constructing executable dependency networks that permit both forward and backward chaining reasoning and renders knowledge dependency networks with a fully executable graphic representation so that networks appear just like they would on a white board.  Because the inference engine is integral, networks can be evaluated in real-time with nodes changing color to indicate their changing "trueness" levels.  This ability to peer into the logical workings of a knowledge network greatly optimizes the knowledge engineering process by

  - providing the ability to run and evaluate freshly elicited knowledge in the presence of the domain expert, 
  - enabling the knowledge engineer to trace the logic structure from data to conclusions and 
  - allowing the knowledge engineer to quickly identify and edit errors and inconsistencies in the logic. 

Knowledge base systems come in a variety of forms, but the dominant type currently in use are rule-based systems. Knowledge representation in NetWeaver, in contrast, is based on object-oriented fuzzy-logic networks which offer several significant advantages over the more traditional rule-based representation. Compared to rule-based knowledge bases, NetWeaver knowledge bases are easier to build, test, and maintain because the underlying object-based representation makes these types of knowledge bases very modular. The modularity of NetWeaver knowledge bases, in turn, allows the designer to gradually evolve complex knowledge bases from simpler ones in small, simple steps. Modularity also allows interactive knowledge base debugging at any and all stages of knowledge base development which expedites the develop process. 

Finally, fuzzy logic provides a formal and complete calculus for knowledge representation that is less arbitrary than the confidence factor approach used in rule-based systems and much more parsimonious than bivalent rules.