计算机科学 ›› 2017, Vol. 44 ›› Issue (Z11): 65-67.doi: 10.11896/j.issn.1002-137X.2017.11A.012
常炳国,李玉琴,冯智超,姚山虎
CHANG Bing-guo, LI Yu-qin, FENG Zhi-chao and YAO Shan-hu
摘要: 运用新一代信息技术快速预测慢性肝病的机理和特征,是提高慢性肝病诊断率的有效途径。运用主成分分析机器学习算法,对描述慢性肝病的多项指标属性项进行降维处理,结合神经网络学习,构建了慢性肝病预测模型。实验分析了125组20维慢性肝病患者的医学检验指标数据项,利用ROC(Receiver Operating Characteristic)曲线优选出13维指标项作为慢性肝病敏感度高的检验指标属性项。通过主成分分析将13维指标项降至5维综合数据项。神经网络训练115组检验指标样本集,剩余10组样本集作为测试样本。与原始20维数据作为神经网络输入相比,所提模型不仅降低了复杂度,且预测精度提高了15.07%。
[1] 世界卫生组织.http://www.who.int/campaigns/hepatitis-day/2016/event/zh. [2] 王恩成,唐琳,王健,等.慢性乙型肝炎中医症候聚类分析研究[J].中国中西医结合杂志,2014,34(1):39-42. [3] HO C Y,LAI Y C,CHEN I W,et al.Statistical Analysis of FalsePositives and False Negatives from Real Traffic with Intrusion Detection/Prevention Systems[J].Communications Magazine,IEEE,2012,50(3):146-154. [4] PAXSON V,ASANOVIC K,DHARMAPURIKAR S,et al.Rethinking hardware support for network analysis and intrusion prevention[C]∥Proc of the 15th USENIX Workshop on Hot Topics in Security.Berkeley,CA:USENIX,2006:63-68. [5] THOMPSON K.Programming techniques:Regular expressionsearch algorithm[J].Communications of the ACM,1968,11(6):419-422. [6] RABIN M O,SCOTT D.Finite automata and their decisionproblems[J].IBM Journal of Research & Development,1959,3(2):114-125. [7] 高秀娟.BP神经网路在肝硬化治疗预测中的应用[J].数学理论与应用,2011,31(3):21-23. [8] 李耐萍.基于机器算法的肝纤维化无创诊断分析与比较[D].长沙:中南大学,2013:30-36. |
No related articles found! |
|