Knowledge Patterns

When building a knowledge base, one frequently repeats similar versions of general theories in multiple, more specific theories. For example, when building the Botany Knowledge Base[1], we embedded a theory of production in representations of photosynthesis, mitosis, growth, and many other botanical processes. Typically, a general theory is incorporated into more specific ones by an inheritance mechanism. However, this works poorly in two situations: when the general theory applies to a specific theory in more than one way, and when only a selected portion of the general theory is applicable. We address this problem with a knowledge engineering technique based on the explicit representation of knowledge patterns, ie. general templates denoting recurring theory schemata, and their transformation (through symbol renaming) to import into specific theories. This technique provides considerable flexibility. A knowledge pattern may be transformed in multiple ways, and each resulting theory can be imported in whole or in part. We describe an application built using this technique, then critique its strengths and weaknesses. We conclude that this technique enables us to better modularize knowledge-bases and to reuse their general theories. 1 The Limitations of Inheritance Consider the following fragment of a hypothetical knowledge-base about computers, expressed in Prolog: % Basic facts about myComputer, an instance of the class of computers: isa(myComputer,computer). speed(myComputer,400). /* MHz */ ram_size(myComputer, 128). /* MB */ disk_space(myComputer,2000) /* MB */ expansion_slots(myComputer,4). % “Available RAM space is the total RAM minus the occupied RAM.” available_ram(Computer,A) :isa(Computer,computer), ram_size(Computer,S), Variables start with upper-case letters and are universally quantified; ‘:-’ denotes reverse implication (←); ‘,’ denotes conjunction; and is denotes arithmetic computation.