How to Understand Belief Drift? Externalization of Variables Considering Different Background Knowledge. (4th December 2018)

• Record Type: Journal Article
• Authors: Hayashi, Teruaki; Ohsawa, Yukio
• Other Names: Rzepka Rafal Guest Editor.
• Is Part Of: Advances in human-computer interaction. Volume 2018(2018)
• Publication Date: 2018-12-04
• Subjects: Human-computer interaction -- Periodicals; Human-computer interaction; Periodicals; Electronic journals; 004.019
• DOI: 10.1155/2018/9054685
• Languages: English
• ISSNs: 1687-5893

Abstract

Abstract : It is necessary to make decisions by integrating appropriate information that is not used in daily life in disaster prevention before, during, and after disasters. Despite this, it is difficult for people to make use of appropriate information under circumstances where various kinds of information are complicated. People can be in an agitated state in which they do not know what will happen. In this paper, we define this situation as Belief Drift (BD) and discuss what kinds of data should be acquired to understand situations of BD because factors causing BD may be diverse. We collected explanations of BD from researchers with different background knowledge and discussed sets of variables inferred by VARIABLE QUEST (VQ). VQ is the inferring method for variables unifying cooccurrence graphs of variables in the datasets. The results indicate that common variables are externalized from the different explanations of BD by researchers with different background knowledge. Results suggest that, even if the terms used to explain the state of BD differ, the data acquired to understand BD are common.