Hierarchical refinement and representation of the causal relation

Abstract

This research looks at the complexity inherent in the causal relation and the implications for its representation in a Terminological Knowledge Base (TKB). Supported by a more general study of semantic relation hierarchies, a hierarchical refinement of the causal relation is proposed. It results from a manual search of a corpus which shows that it efficiently captures and formalizes variations expressed in text. The feasibility of determining such categorization during automatic extraction from corpora is also explored. Conceptual graphs are used as a representation formalism to which we have added certainty information to capture the degree of certainty surrounding the interaction between two terms involved in a causal relation.

Extracting Explicit and Implicit Causal Relations from Sparse, Domain-Specific Texts

“…"breaker voltage exceeded allowable limit"). We observed that these relations had optional causal-agents, and that they could not be conclusively classified in Barriere's taxonomy of causality [1]. When the causal-agents were specified, as in "[short-circuit in brake wiring causal-agent ] destroyed the power supply", the relations indicated the creation or destruction of events.…”

“…These results also demonstrate that Wikipedia can be effectively leveraged upon as a knowledge-base for extracting domain-specific causal relations. Furthermore, we found out that some of the identified domain-specific relations were not conclusively classifiable in Barriere's taxonomy of causality [1].…”

“…A recent taxonomy of causal relations is that of Barriere [1], which distinguishes between existential and influential causality. Existential causality pertains to the creation, destruction, prevention and maintenance of events (or entities).…”

“…We implemented our methodology in a prototype as part of the DataFusion initiative 1 , which aims at enhancing product quality and customer satisfaction using information extracted from corporate texts. The crux of our approach lies in acquiring a set of explicit and implicit causal patterns from Wikipedia, which we exploit as a knowledge-base.…”

Semantic relations in information science

“…Other classifications of the part-whole relation have been developed by Barriere (1997Barriere ( & 2002, Markowitz, Nutter & Evens (1992), Sattler (1995) specifically for an engineering application, Uschold (1996) for ecological information systems, and Bernauer (1996) for the medical domain.…”