Research on Question Answering System for COVID-19 Based on Knowledge Graph
Kai Ding, Hongqi Han, Linna Li, Menglin Yi · 2021
In the context of the COVID-19 epidemic, it is necessary to establish a question answering (QA) system to help people to get information about the epidemic. This paper creats a framework about the question answering system based on knowledge bases (KBQA) for COVID-19. First, we design the schema of the COVID-19 knowledge graph (KG) and extract knowledge from texts. Second, a rule-based classifier is created to find the user’s intention when they input a question. Then, the matched template is used to transfer the question into Cypher query, and return answer which is retrieved from KG to users. We collected data from Baidu Encyclopedia and extracted triplet knowledge for constructing COVID-19 KG to prove the feasibility of our framework. The experiment on 86 questions shows that the KBQA for COVID-19 can achieve 81.6% accuracy and get a good experience through the feedback of testers.