基于SimRank++算法和协同过滤算法的疾病-症状关联性研究  

Research on Correlation Between Diseases and Symptoms Based on SimRank++and Collaborative Filtering

作　　者：马健 MA Jian(College of Electronic and Information Engineering,Tongji University,Shanghai 201804,China)

机构地区：[1]同济大学电子与信息工程学院,上海201804

出　　处：《信息与电脑》2023年第2期60-63,共4页Information & Computer

摘　　要：文章基于临床电子病历中医疗实体的共现关系,先构建疾病-症状二分网络,再利用SimRank++算法和协同过滤算法共同实现疾病实体和症状实体之间相关关系的度量,最后应用Spark分布式平台来适应大量数据的高计算强度,以提高计算效率。实验表明,本文方法可以有效度量疾病实体和症状实体之间的相关程度,可为临床医生的疾病诊断提供帮助。Based on the co-occurrence relationship of medical entities in clinical electronic medical records,this paper constructs a disease-symptom binary network,and uses SimRank++algorithm as well as collaborative filtering algorithm to jointly measure the correlation between diseases and symptoms.In addition,the distributed platform Spark is used to adapt to the huge computation of large amounts of data to improve efficiency.The experiment shows that the method in this paper can effectively measure the correlation between diseases and symptoms,and provide help for clinicians in disease diagnosis.

关 键 词：电子病历 SimRank++ 协同过滤 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]