Memory-Based Hypothesis Formation: Heuristic Learning of Commonsense Causal Relations from Text

We present a memory-based approach to learning commonsense causal relations from episodic text. The method relies on dynamic memory that consists of events, event schemata, episodes, causal heuristics, and cousol hypotheses. The learning algorithms are based on applying causal heuristicsto precedents of new information. The heuristics are derived from principles of causation, and, to a limited extent, from domain-related causal reasoning. learning is defined as finding--and later augmenting-inter-episodal and intro-episodal causal connections. The learning algorithms enable inductive generalization df causal associations into AND/OR graphs. The methodology has been implemented and tested in the program NEXUS.