A Net Structure Based Relational Question Answerer: Description and Examples

Abstract

A question answering system is described which uses a net structure for storage of information. The net structure consists of nodes and labelled edges, which represent relations between the nodes. The labels are also nodes, and therefore definitions of relations may be stored in the net. It is demonstrated that the generality and complexity of this memory structure allows a surprisingly powerful question answering system to be constructed using comparitively simple executive routines. Output from the question answerer, which is currently running on an interactive, time sharing system, is included, showing its range of applicability including question answering, inductive and deductive inference, simple theorem proving and problem solving. Key words and phrases: relational question answering, question answering, memory net, memory structure, data structure, semantic memory, semantic information retrieval, deductive inference, inductive inference, problem solving, concept formation, relational logic, learning, theorem proving, fact retrieval.