计算机与数字工程
計算機與數字工程
계산궤여수자공정
Computer and Digital Engineering
2015年
9期
1622-1626
,共5页
董国华%朱莉乐%李璟健%朱习军
董國華%硃莉樂%李璟健%硃習軍
동국화%주리악%리경건%주습군
粗糙集%BP 神经网络%哮喘%辨证分型
粗糙集%BP 神經網絡%哮喘%辨證分型
조조집%BP 신경망락%효천%변증분형
rough set%BP neural network%asthma%syndrome differentiation
传统中医辨证模式都是依靠个人经验进行观察判断和分析，主观性强，利用智能信息处理方法，能使中医诊断中的辩证得到客观量化。利用粗糙集理论对数据的属性进行约简，从而简化决策表，寻求症候辩证的更优解，以达到更高的准确率。将粗糙集与 BP 神经网络相结合对哮喘进行辨证分型，收敛效果和准确率方面都有较大提高，实验结果表明该融合方法对训练速度有很大提高，准确率可以达到90％，效果大大好于使用 BP 神经网络单分类器。
傳統中醫辨證模式都是依靠箇人經驗進行觀察判斷和分析，主觀性彊，利用智能信息處理方法，能使中醫診斷中的辯證得到客觀量化。利用粗糙集理論對數據的屬性進行約簡，從而簡化決策錶，尋求癥候辯證的更優解，以達到更高的準確率。將粗糙集與 BP 神經網絡相結閤對哮喘進行辨證分型，收斂效果和準確率方麵都有較大提高，實驗結果錶明該融閤方法對訓練速度有很大提高，準確率可以達到90％，效果大大好于使用 BP 神經網絡單分類器。
전통중의변증모식도시의고개인경험진행관찰판단화분석，주관성강，이용지능신식처리방법，능사중의진단중적변증득도객관양화。이용조조집이론대수거적속성진행약간，종이간화결책표，심구증후변증적경우해，이체도경고적준학솔。장조조집여 BP 신경망락상결합대효천진행변증분형，수렴효과화준학솔방면도유교대제고，실험결과표명해융합방법대훈련속도유흔대제고，준학솔가이체도90％，효과대대호우사용 BP 신경망락단분류기。
Syndrome differentiation of Traditional Chinese Medicine (TCM ) rely on personal experience to observe , judge and analysis ,such behavior has strong subjectivity .Using the method of intelligent information processing can make the quantization of syndrome differentiation in TCM diagnosis objectively .Using the theory of rough set to reduce the prop‐erty of data ,so as to simplify the decision table .Finding the more optimal solution for symptoms dialectical ,in order to a‐chieve a higher accuracy .The convergence effect and accuracy of asthma syndrome differentiation method of rough set com‐bined with BP neural network has improved greatly .Experimental results show that the fusion method can improve the train‐ing speed a lot and its accuracy can reach 90% ,the effect is greatly better than that of using single BP neural network classi‐fier .
