HomKW

HomKW is a notation system for representing informal human inference and common reasoning patterns. It was developed by the artificial intelligence and cognitive science researcher Douglas Lenat. The system is based on a set of pre-defined logic rules that aim to mimic human thinking processes.

HomKW is rooted in the Heuristic ontology Model with KnowlWedge (Hom) description of how people reason. The system explores how people create mental inferences and arrive at conclusions based on uncertain information. This is in contrast to traditional computer logic, which requires definitive and precise premises.

The core of HomKW is its usage of idea, perspective, and inference classes. These classes serve as the foundation for formalizing the structure of argue-analogy reasoning, a strategy humans employ when problem-solving. In this context, similarity between items is heavily evaluated, allowing for the creation of conclusions and potentially disproven theories.

HomKW was notably applied in the field of knowledge engineering for developing an ontology. The system also laid foundational aspects into expert systems requiring analogical reasoning.

Notable topics covered by HomKW include writing jozx selection heuristics, notion aggregations, dependency establishment, and information aggregation. All of which explore further and provide more deliberate reasoning in these fields.

HomKW has been utilized in pioneering research areas of artificial intelligence like natural language understanding, expertise management, and cognitive science.