Understanding a story with causal relationships

Abstract

In our previous work (e.g., [4], [5], [6], [7]), we have formalized the story understanding process based on scripts and plans with stepwise default theories. While those theories offer final results for understanding a specific story, they do not provide the history of changes of partial states of any objects the story may concern. Moreover, the causal models for missing events are incomplete in script-based understanding, and even not involved in plan-based understanding. As the result, the understanding process lacks the causal foundation. In this paper, we propose a default rule representation of causal relationships. In common sense situations, we give an event-based analysis for this general representation to fix its structure. A complete causal model for a story, i.e., a default causal chain, is developed for understanding the story. Stepwise default theories and frame-based systems are described. The latter provides the history of partial state changes of agents and objects in the story by generating an understanding chain.

The author is supported by Sino-British Friendship Scholarship Scheme (SBFSS) and Natural Science Foundation of China (NSFC).