摘要
本课题从深度学习的角度研究中药处方的配伍规律,通过深层神经网络捕获处方中药材搭配组合规律。实验实现对处方及其中药材的机器评价,并发现处方中潜在不恰当药材评价具有显著低于处方总体评价的特性。利用该特性进行处方审查,有望能够有效提升审查环节效率和准确性。
This project studies the compatibility of Chinese prescriptions from the perspective of deep learning,and captures the pattern of prescription herbs matching and combination through deep neural networks.The machine evaluation of prescriptions and their herbs was achieved,and it was found that the evaluation of potentially inappropriate herbs in prescriptions was significantly lower than the overall evaluation of prescriptions.Using this feature to conduct prescription reviews is expected to improve the review process’s efficiency and accuracy effectively.
作者
蒲翊凡
李婵娟
杨可心
PU Yifan;LI Chanjuan;YANG Kexin(College of Computer Science,Sichuan University,Chengdu 610065)
出处
《现代计算机》
2021年第16期55-58,共4页
Modern Computer
关键词
处方审查
中药配伍
机器评价
Prescription Review
Compatibility of Traditional Chinese Medicine
Machine Evaluation
