Author(s)

Zhuoran Zhen, Dianhui Chu, Chunshan Li, Yuanyuan Wang

Corresponding Author

Zhuoran Zhen

ABSTRACT

Nowadays, population aging has become a prominent problem all over the world, which brings new challenge to manage huge health data. Existing approaches store those huge health knowledge data in traditional relational databases through a labor-intensive process and lack well-defined ontology to describe entities and relations in elderly health domain. Knowledge graph has better knowledge managing ability than traditional storage methods due to its logical structure. In this paper, we design a knowledge graph (KnowHealth) to manage the elderly’s health data. Specifically, we automatically crawl and extract information related to health from different input sources, and construct a knowledge graph to manage them in the graph database. We give an ontology definition which describes the entities and relations with different types and attributes. In order to serve query for the elderly, we design and implement a historical behavior driven by a question-answer platform. Finally, we conduct a discussion on several key technologies for KnowHealth.

KEYWORDS

Knowledge Graph, Biomedical Text Mining, Relation Extraction, Question Answering Platform