基于隐语义模型的中医在线辅助诊疗系统

发布时间：2018-06-22 18:40

  本文选题：数据挖掘 + 隐含狄利克雷分布　； 参考：《计算机应用》2017年S1期

【摘要】：当前对中医学的怀疑关键在于其缺少科学数据的支撑,因此,把中医诊疗的过程数据化十分重要。针对该问题提出一种数据驱动的中医诊疗方法,基于对医案中病症和对应处方的隐语义分析,找出隐含病机,发现隐含病机与病症和药物间存在的关系,建立了一个基于传统中医医案挖掘的多内容隐含狄利克雷分布(LDA)模型。基于模型的结果,提出根据症状推荐药物的算法,并且建立了基于隐语义模型的中医在线辅助诊疗系统。通过实验评估推荐算法的有效性,在精度、召回率方面均好于基线方法。中医在线辅助诊疗系统能提供数据驱动的诊疗结果辅助中医师诊疗,帮助中医更准确、全面、智能地制定科学的治疗方案。
[Abstract]:At present, the key to the doubt of TCM lies in its lack of scientific data, so it is very important to digitize the process of TCM diagnosis and treatment. To solve this problem, a data-driven method of TCM diagnosis and treatment was proposed. Based on the implicit semantic analysis of symptoms and corresponding prescriptions in medical records, the underlying pathogenesis was found, and the relationship between hidden pathogenesis and symptoms and drugs was found. A multi-content implicit Delikley distribution (LDA) model based on traditional Chinese medicine case mining is established. Based on the results of the model, an algorithm for recommending drugs according to symptoms is proposed, and an online Chinese medicine diagnosis and treatment system based on implicit semantic model is established. The effectiveness of the recommendation algorithm is evaluated experimentally, and the accuracy and recall rate are better than the baseline method. Chinese medicine online assistant diagnosis and treatment system can provide data-driven diagnosis and treatment results to assist traditional Chinese medicine doctors to make more accurate, comprehensive and intelligent scientific treatment plan.
，

本文编号：2053816