Premium
Memory‐Based Hypothesis Formation: Heuristic Learning of Commonsense Causal Relations from Text
Author(s) -
Bozsahin H. Cem,
Findler Nicholas V.
Publication year - 1992
Publication title -
cognitive science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.498
H-Index - 114
eISSN - 1551-6709
pISSN - 0364-0213
DOI - 10.1207/s15516709cog1604_1
Subject(s) - heuristics , computer science , artificial intelligence , heuristic , generalization , causality (physics) , causal model , commonsense reasoning , causal reasoning , causation , cognitive science , machine learning , commonsense knowledge , cognitive psychology , natural language processing , psychology , cognition , domain knowledge , mathematics , epistemology , philosophy , statistics , physics , quantum mechanics , neuroscience , operating system , mathematical analysis
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 causal hypotheses. The learning algorithms are based on applying causal heuristics to 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 intea‐episodal causal connections. The learning algorithms enable inductive generalization of causal associations into AND/OR graphs. The methodology has been implemented and tested in the program NEXUS.