吉首大学学报(自然科学版)

• 计算机 • 上一篇    下一篇

疾病症状与类型关系的苗医药大数据挖掘方法

杨博,黄斌   

  1. (1.贵州师范大学数学科学学院,贵州 贵阳 550001;2.贵州师范大学大数据与计算机科学学院,贵州 贵阳 550001)
  • 出版日期:2017-11-25 发布日期:2017-12-29
  • 通讯作者: 黄斌(1971-),男,湖南怀化人,教授,博士,主要从事大数据、云计算、认知智能研究.
  • 作者简介:杨博(1989-),男,山西太原人,硕士研究生,主要从事大数据研究
  • 基金资助:

    国家自然科学基金项目(61540052);贵州省科技创新人才团队建设项目 ([2016]5629)

Miningmethod of symptoms and disease relationship based on big data in Miao medicine

YANG Bo,HUANG Bin   

  1. (1.Guizhou Normal University,School of Mathematical Sciences,Guizhou Guiyang 550001;2.Guizhou Normal University,School of Big Data and Computer Science,Guizhou Guiyang 550001)
  • Online:2017-11-25 Published:2017-12-29

摘要:

苗族文字的缺失造成了苗药药效记录的缺失,从而影响苗药的规范研制和科学用药.苗医药大数据挖掘通过分析病历中疾病与用药之间的关系挖掘出苗药药效,为苗药药效记录的缺失问题提供了新的解决思路.然而由于各地苗医的技术水平参差不齐,苗医在分辨疾病时会出现偏差,这给从苗医药大数据中挖掘出苗药药效造成了困难.针对病历数据中疾病分类存在偏差的问题,本研究提出了一个基于分布式k近邻的、面向疾病症状与类型关系的苗医药大数据挖掘方法.实验结果表明在数据量大的情况下该方法具有较高的分类准确度.

关键词: 苗医药, 大数据, 症状, 疾病, MapReduce, Hadoop

Abstract:

The lack of Miao characters leads to the lack of efficacy of Miao medicine,so as to affect the development and scientific use of Miao medicine.The Big Data in Miao medicine provides a new way to solve the problem of lack of drug properties.Efficacy of Miao medicine can be found by analyzing the relationship between disease and drug use in medical records.However,because of poor medical technology,some Miao-nationality doctors gave wrong disease when they handled some cases.As wrong disease,it is difficult to mine the efficacy of Miao medicine from the Big Data in Miao medicine.For wrong disease,this paper presents a mining method of Miao medicine symptoms and disease based on distributed k nearest neighbor.In the case of large amount of data,the method has good classification performance and high accuracy.

Key words: Miao medicine, big data, symptom, disease, MapReduce, Hadoop

公众号 电子书橱 超星期刊 手机浏览 在线QQ