Research On Relation Search Techniques Based On Disease Knowledge Graph

In recent years,more progress and achievement on health-care tech enable people to acquire more data about biology and medicine.And with the rise of precision medicine it has become a hot discussion about how to utilize data generated from biology and medicine to promote the level of modern medicine and health care.In various kinds of biological data,the data about disease is the closest to people’s health condition.It is of vital importance to reach a higher standard of disease study and to enhance people’s health quality if we can analyze the relation among diseases better.According to situations mentioned above,this paper analyses the relation of several data related to disease and try to use different kind of method to study the searching techniques on the knowledge network constructed by disease data.In the first part of this paper,we studied different kind of disease-related data and the knowledge graph.Then we referred to the advantage of knowledge graph on relation computing and knowledge expression and convert the disease data into the ontologies in knowledge graph.After that we transformed the expression model of the data,which made it more convenient for the realization of algorithms.According to the features of our data used in this paper we studied some algorithms,especially the algorithms in search engine and recommender systems,which are designed to discover the relations among massive data,and we referred to some models to implemented two algorithms to achieve the relation search process on disease knowledge graph.After the implementation of the two algorithms,we realized some improvements to overcome the time-consuming problems appeared in both algorithms and promote the efficiency of themAt the last of the paper,we compared the two algorithms with several kinds of evaluation methods and the output show different features of the them.Besides,We tested the two algorithms about their improvement on running speed,which can tell they performed better after improved.