Construction of a Causal-Chain Presentation System Considering Output Diversity for the Discovery of Economic Ripple Effects

Abstract

External events, like epidemics or rising material costs, significantly impact companies' activities. Policymakers need to understand the ripple effects of these events to identify companies in need of assistance. In this study, we construct a system that presents a chain of other economic events derived from a given event by extracting descriptions of causal relationships from a large amount of textual data. This allows for the discovery of economic ripple effects to aid decision-making. The objective of this study is to construct a causal-chain presentation system considering output diversity to discover a wider range of economic ripple effects. To achieve this objective, we developed a new algorithm utilizing Maximal Marginal Relevance to present causal chains. Through evaluation experiments using financial statement summaries, we confirmed that our approach produces "more diverse output without sacrificing accuracy." Furthermore, we analyzed a case study to validate the benefits of considering output diversity.