Information Retrieval System and Knowledge Base on Diseases Using Variables and Contexts in the Texts
Teruaki Hayashi, Yukio Ohsawa · Procedia Computer Science · 2019
In the field of disaster prevention, it is crucial to provide the general public with appropriate integrated information before, during, and after the disasters. However, in such situations, individuals are not capable of establishing clear opinions owing to inconsistent and diverse information from multiple sources. As individuals receive varied inconsistent information from several sources, they are puzzled about the credibility of such information. This situation is defined as a belief drift in our study, and it must be alleviated by providing appropriate information. To create a system that provides the required information to individuals with drifting beliefs, we acquired reliable information from medical doctors and implemented the knowledge base and retrieval system to bridge the gap between professionals and nonprofessionals. We focused on certain blood diseases, collected large texts, and extracted the terms representing contexts and variables, and the medical doctors evaluated the relationships among these diseases. In this study, we discussed the design and the implementation of the proposed system and demonstrated its performance.