神经网络集成研究及其在PTA生产中的应用
|
摘要
近年来,神经网络集成技术已成为机器学习领域研究的热点之一,它可以获得比单个神经网络更好的泛化能力和稳定性,其中的选择性集成技术由于其在推广性和组合性方面的优势得到了广泛关注。
PTA(精对苯二甲酸)作为主要的化工原料,涉及国民经济的各个方面,应用领域非常广泛,因此研究PTA生产过程具有重要的实际意义。为解决PTA流程工业中的复杂模式和工作点漂移引起的单个神经网络模型泛化能力不足和稳定性问题,本文提出一种新的基于权值信息度量网络差异度的选择性神经网络集成方法DWSEN。DWSEN利用各个子网络对应的权值去度量网络模型之间的差异性,并给出用于计算相同结构网络之间差异度的计算公式,对比传统的集成方法和目前流行的神经网络集成算法Bagging和Boosting,仿真实验结果显示DWSEN具有更好的泛化能力和更强的稳定性能。将DWSEN方法用于PTA工业中薄膜蒸发器蒸发量和溶剂脱水塔塔顶电导率的建模过程,所得集成模型能够提高PTA生产装置的软测量模型的精度和稳定性,能更好的反映蒸发量和电导率的变化趋势,本方法为工业过程的复杂机理模型分析提供了一种行之有效的建模方案。
Recently, neural network ensemble has become one of hot topics of machine learning technology. And it has the advantages of generalization ability and stability than a single artificial neural network. Selective ensemble technology received extensive attentions because of its advantages of applicability and combinability.
Purified terephthalic acid (PTA), as a major chemical raw material, covering all aspects of the national economy, has been applied widely in various fields. Thus, the study of PTA production process has great practical significance. In order to solve the lack of generalization and stability in single neural network model because of the complexity of mechanism and operating point drift, the paper puts forward a new selective neural network ensemble approach named DWSEN which to measure the diversity of individuals according to weights of individual networks. And it presents the formula used to calculate the diversity of networks with same network structure. Comparing with some prevailing ensemble approaches such as Bagging and Boosting, experiments reflect that DWSEN has higher generalization ability and stronger stability. The method is applied to the modeling of evaporation of Film Evaporator and tower conductivity of Solvent Dehydration Tower in PTA industrial process. Cases study show that the obtained models can possess better generalization performance, and simulate the production process better in PTA industry. DWSEN provides a well-estabilished modeling program for complex mechanism model in PTA process industry.
PTA(精对苯二甲酸)作为主要的化工原料,涉及国民经济的各个方面,应用领域非常广泛,因此研究PTA生产过程具有重要的实际意义。为解决PTA流程工业中的复杂模式和工作点漂移引起的单个神经网络模型泛化能力不足和稳定性问题,本文提出一种新的基于权值信息度量网络差异度的选择性神经网络集成方法DWSEN。DWSEN利用各个子网络对应的权值去度量网络模型之间的差异性,并给出用于计算相同结构网络之间差异度的计算公式,对比传统的集成方法和目前流行的神经网络集成算法Bagging和Boosting,仿真实验结果显示DWSEN具有更好的泛化能力和更强的稳定性能。将DWSEN方法用于PTA工业中薄膜蒸发器蒸发量和溶剂脱水塔塔顶电导率的建模过程,所得集成模型能够提高PTA生产装置的软测量模型的精度和稳定性,能更好的反映蒸发量和电导率的变化趋势,本方法为工业过程的复杂机理模型分析提供了一种行之有效的建模方案。
Recently, neural network ensemble has become one of hot topics of machine learning technology. And it has the advantages of generalization ability and stability than a single artificial neural network. Selective ensemble technology received extensive attentions because of its advantages of applicability and combinability.
Purified terephthalic acid (PTA), as a major chemical raw material, covering all aspects of the national economy, has been applied widely in various fields. Thus, the study of PTA production process has great practical significance. In order to solve the lack of generalization and stability in single neural network model because of the complexity of mechanism and operating point drift, the paper puts forward a new selective neural network ensemble approach named DWSEN which to measure the diversity of individuals according to weights of individual networks. And it presents the formula used to calculate the diversity of networks with same network structure. Comparing with some prevailing ensemble approaches such as Bagging and Boosting, experiments reflect that DWSEN has higher generalization ability and stronger stability. The method is applied to the modeling of evaporation of Film Evaporator and tower conductivity of Solvent Dehydration Tower in PTA industrial process. Cases study show that the obtained models can possess better generalization performance, and simulate the production process better in PTA industry. DWSEN provides a well-estabilished modeling program for complex mechanism model in PTA process industry.
引文
[1]李涛.PTA工业生产技术现状及其对策研究[J].合成纤维,2004,1:10-12
[2]章文.2003-2007年中国精对苯二甲酸扩建情况[J].上海化工,2003,3:45-46
[3]张林春.国内外PTA生产现状和及国内发展前景[J].金山油化纤,2003,22(3):24-26
[4]袁增任.人工神经元网络及其应用[M].北京:清华大学出版社,2003:66-77
[5]McCulloch W S,Pitts W A.Logical calculus of the ideas immanent in nervous activity[J].Bulletin of Mathematics and Biophysics,1943,5:115-133
[6]郭鑫.广义回归神经网络和遗传算法研究及其在化工过程建模中的应用[D].杭州:浙江大学化工计算机仿真与系统工程专业,2004
[7]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京:清华大学出版社,2005:184-195
[8]Rosenblatt F.The Perceptron:A probabilistic model for information storage and organization in the brain[J].Psychological Review,1958,65(6):386-408
[9]Hopfield J J.Neural networks and physical systems with emergent collective computational abilities[C].Proceedings of the National Academy of Sciences,USA,1982,79:2554-2558
[10]Rumehart D E,Hinton G E,Williams R J.Learning internal representations by error propagation in Parallel Distributed processing[J].MA:MIT Press,Cambridge,1986:318-362
[11]Campos M,Lopez D.Neural network approach to locate motifs in biosequences[J].Proceedings of Progress in Pattern Recongnition,Image Analysis and Applications,2005,3773:214-221
[12]Armano G,Mancosu G,Milanesi L,Orro A,Saba M,Vargiu E.A hybrid genetic-neural system for predicting protein secondary structure[J].Bmc Bioinformatics,2005
[13]Kato R,Nakano H,Konishi H,Kato K,Koga Y,Yamane T,Kobayashi T,Honda H.Novel strategy for protein exploration:High-throughput screening assisted with fuzzy neural network[J].Journal of Molecular Biology 2005,351:683-692
[14]Coban R,Can B.Identification and control of ITU Triga Mark-Ⅱ Nuclear Research Reactor using neural networks and fuzzy logic[J].Advances in Artificial Intelligence,2005,3809:1057-1062
[15]Guo H J,Lin S F,Liu J H.A radial basis function sliding mode controller for chaotic Lorenz system[J].Physics Letters,2006,351:257-261
[16]Pai P F,Lin K P.Application of hybrid learning neural fuzzy systems in reliability predicton[J].Quality and Reliability Engineering International,2006,22:199-211
[17]Liu Y G,Li Y M.Sliding mode adaptive neural-network control for nonholonomie mobile modular manipulators[J].Journal of Intelligent & Robotic Systems,2005,44:203-224
[18]Yildirim S.Design of adaptive robot control system using recurrent neural network[J].Journal of Intelligent & Robotic Systems,2005,44:247-261
[19]Hansen L K,Salamon P.Neural network ensembles[J].IEEE Trans Pattern Analysis and Machine Intelligence,1990,12(10):993-1001
[20]Sollich P,Krogh A.Learning with ensembles:how over-fitting can be useful[J].Advances in Neural Information Processing Systems 8,Denver,CO:MIT Press,Cambridge,MA,1996:190-196
[21]Hansen L K,Liisberg L,Salamon P.Ensemble methods for handwritten digit recognition[J].In Proceedings of IEEE Workshop on Neural Networks for Signal Processing,Copenhagen,Denmark,1992:333-342
[22]Hampshire J,Waibel A.A novel objective function for improved phoneme recognition using timedelay neural networks[J].IEEE Transactions on Neural Networks,1990,1(2):216-228
[23]Gutta S,Wechsler H.Face recognition using hybrid classifier systems[J].In proceedings of IEEE International Joint Conference on Neural Network,Washington,1996:1017-1022
[24]Sharkey A J C,Sharkey N E,Cross S S.Adapting an ensemble approach for the diagnosis of breast cancer[J].In proceedings of International Conference on Artificial Neural Networks,Skovde,Sweden,1998:281-286
[25]Schapire R E,Singfer Y.BoosTexter:A boosting-based system for text categorization[J].Machine Learning,2000,39(2-3):135-168
[26]Dietterich T G.Ensemble methods in machine learning[J].In:Multiple Classifier Systems:Springer,2001:1-15
[27]周志华,陈世福.神经网络集成计算机学报[J].2002,25(1):1-8
[28]Valiant G.Atheory of the learnable[J].Communications of the ACM,1984,27(11):1134-1142
[29]Kearns M,Valiant G.Learning Boolean formulae or factoring[R].Technical Report TR-1488,Cambridge,MA:Harvard University Aiken Computation Laboratory,1988
[30]Krogh A,Vedelsby J.Neural network ensembles,cross validation,and active learning[J].In:Advances in Neural Information Processing Systems 7,CO:MIT Press,Cambridge,MA,1995:231-238
[31]傅强.选择性神经网络集成算法研究[D].浙江:浙江大学控制理论与工程,2007
[32]沈学华,周志华,吴建鑫,陈兆乾.Boosting和Bagging综述[J].计算机工程与应用,2000,12:31-33
[33]董乐红,耿国华,高原.Boosting算法综述[J].计算机应用与软件,2006,23(8):27-29
[34]Schapire R E.The strength of weak learnability[J].Machine Learning,1990,5(2):197-227
[35]Freund Y.Boosting a weak algorithm by majority[J].Information and Computation,1995,121(2):256-285
[36]Breiman L.Bagging predictors[J].Machine Learning,1996,24(2):123-140
[37]王正群,陈世福,陈兆乾.基于模糊划分的神经网络集成[J].南京大学学报(自然科学版),2006,42(1):63-67
[38]乐晓蓉,王正群,郭亚琴,王向东.基于聚类算法的神经网络集成[J].计算机工程,2007,33(5):180-184
[39]吕强,刘士荣,俞金寿.神经网络聚类加权集成及其应用[J].中南大学学报(自然科学版),2007,38(sup.l):32-35
[40]凌锦江,陈兆乾,周志华.基于特征选择的神经网络集成方法[J].复旦学报(自然科学版),2004,43(5):685-688
[41]凌锦江,周志华.基于因果发现的神经网络集成方法[J].软件学报,2004,15(10):1479-1484
[42]刘悦,吴耿锋,丁智国.基于特征加权的神经网络集成及其应用[J].上海大学学报(自然科学版),2007,13(5):623-628
[43]Lazarevic A,Kumar V.Feature bagging for outliner detection[J].KDD,Chicago,Illinois,USA,2005:157-166
[44]Partridge D,Yates W B.Engineering multiversion neural-net systems[J].Neural Computation,1996,8(4):869-893
[45]Maclin R,Shavlik J W.Combining the predictions of multiple classifiers:using competitive learning to initialize neural networks[J].Proceedings of the ~(14)th International Joint conference on Artificial Intelligence,Montreal,Canada,1995:524-530
[46]Optiz D W,Shavlik J W.Actively searching for an effective neural network ensemble[J].Connection Science,1996b,8(3-4):337-353
[47]巩文科,李长河,石争浩,赵洁.基于样本重构的神经网络集成学习方法[J].计算机应用,2006,26(6):1428-1430
[48]Islam M M,Yao X,Murase K.A constructive algorithm for training cooperative neural network ensembles[J].IEEE Transactions on Neural Netwoks,2003,14(4):820-834
[49]Liu Y,Yao X.Negatively correlated neural networks for classification[J].Proceedings of the 3~(rd)International Symposium on Artificial Life and Robotics,Beppu,Japan,1998:736-739
[50]Yong L,Xin Y.Simultaneous training of negatively correlated neural networks in ensemble systems[J].IEEE Transactions on Man and Cybernetics,Part B,1999,29(6):716-725
[51]Jang M,Cho S.Observational learning algorithm for an ensemble of neural networks[J].Pattern Analysis & Applications,2002,5(2):154-167
[52]王正群,陈世福,陈兆乾.并行学习神经网络集成方法.计算机学报[J].2005a,28(3):402-408
[53]王正群,陈世福,陈兆乾.一种主动学习神经网络集成方法.计算机研究与发展[J].2005b,42(3):375-380
[54]吴建鑫,周志华,沈学华,陈兆乾.一种选择性神经网络集成构造方法[J].计算机研究与发展,2000,37(9):1039-1044
[55]Zhihua Z,Jianxin W,Wei T.Ensembling neural networks:Many could be better than all[J].Artificial Intelligence,2002,137(122):239-263
[56]Lazarevic A,Obradoric Z.Effective pruning of neural network classifer ensembles[J].Processing International Joint conference on Neural Networks,2001:796-801
[57]Giacinto G,Roli F.Design of effective neural network ensembles for image classification purposes[J].Image and Vision Computing,2001,19:699-707
[58]Bakker B,Heskes T.Clustering ensembles of neural network models[J].Neural Networks,2003,16:261-269
[59]李凯,黄厚宽.一种基于聚类技术的选择性神经网络集成方法[J].计算机研究与发展,2005,42(4):594-598
[60]傅强,胡上序,赵胜颖.基于PSO算法的神经网络集成构造方法[J].浙江大学学报(工学版),2004,38(12):1596-1600
[61]凌霄汉,吉根林.一种快速选择性神经网络集成方法[J].郑州大学学报(理学版),2006,38(4):69-73
[62]施彦,黄聪明,侯朝桢.基于随机梯度法的选择性神经网络二次集成[J].计算机研究与发展2005,42(4):594-598
[63]傅向华,冯博琴,马兆丰,何明.增量构造负相关异构神经网络集成的方法[J].西安交通大学学报,2004,38(8):796-799
[64]陈国华.分层随机抽样方法与小子样检验[J].电子产品可靠性与环境试验,2001,5:11-13
[65]邓明.基于分层随机抽样的季节指数的抽样估计研究[J].统计研究,2008,25(7):70-75
[66]李洁,高新波,焦李成.一种基于修正模糊度的聚类有效性函数[J].系统工程与电子技术,2005,27(4):723-726
[67]诸克军,苏顺华,黎金玲.模糊c均值中的最优聚类和最佳聚类数[J].系统工程理论与实践,2005,25(3):52-61
[68]于剑,程乾生.模糊聚类方法中的最佳聚类数的搜索范围[J].中国科学(E辑),2002,32(2):274-280
[69]高新波,谢维信.模糊聚类理论发展及应用的研究进展[J].科学通报,1999,44(21):2241-2251
[70]Hecht-Nielson R.Theory of the back propagation neural networks[J].Neural Networks for perception,1992,2:65-93
[71]White H.Commentionist nonparametric regression:Multilayer feed forward networks can learn arbitrary mapping[J].Neural Networks,1990,3(4):535-549
[72]张振亚,王进,程红梅,王煦法.基于余弦相似度的文本空间索引方法研究[J].计算机科学,2005,32(9):160-163
[73]广凯,潘金贵.一种基于向量夹角的k近邻多标记文本分类算法[J].计算机科学,2008,35(4):205-207
[74]李希强,陆舍铭,王笛,刘春波,刘惠民,董学畅,缪明明.烟用香精UPLC指纹图谱的夹角余弦、相关系数与欧式距离评价[J].烟草科技,2008,12:37-41
[75]李俊峰,戴文战.基于夹角余弦的图像边缘检测算法及应用[J].中南大学学报(自然科学版),2007,38(sup.1):608-613
[76]C.L.Blake,C.J.Merz.UCI repository of machine learning databases[EB/OL].Http://www.ics.uci.edu/～mlearn/ML Repository.html.2007.12.12
[77]尚海燕.PTA残渣系统与溶剂系统建模[D].北京:北京化工大学控制理论与控制工程专业,2007
[2]章文.2003-2007年中国精对苯二甲酸扩建情况[J].上海化工,2003,3:45-46
[3]张林春.国内外PTA生产现状和及国内发展前景[J].金山油化纤,2003,22(3):24-26
[4]袁增任.人工神经元网络及其应用[M].北京:清华大学出版社,2003:66-77
[5]McCulloch W S,Pitts W A.Logical calculus of the ideas immanent in nervous activity[J].Bulletin of Mathematics and Biophysics,1943,5:115-133
[6]郭鑫.广义回归神经网络和遗传算法研究及其在化工过程建模中的应用[D].杭州:浙江大学化工计算机仿真与系统工程专业,2004
[7]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京:清华大学出版社,2005:184-195
[8]Rosenblatt F.The Perceptron:A probabilistic model for information storage and organization in the brain[J].Psychological Review,1958,65(6):386-408
[9]Hopfield J J.Neural networks and physical systems with emergent collective computational abilities[C].Proceedings of the National Academy of Sciences,USA,1982,79:2554-2558
[10]Rumehart D E,Hinton G E,Williams R J.Learning internal representations by error propagation in Parallel Distributed processing[J].MA:MIT Press,Cambridge,1986:318-362
[11]Campos M,Lopez D.Neural network approach to locate motifs in biosequences[J].Proceedings of Progress in Pattern Recongnition,Image Analysis and Applications,2005,3773:214-221
[12]Armano G,Mancosu G,Milanesi L,Orro A,Saba M,Vargiu E.A hybrid genetic-neural system for predicting protein secondary structure[J].Bmc Bioinformatics,2005
[13]Kato R,Nakano H,Konishi H,Kato K,Koga Y,Yamane T,Kobayashi T,Honda H.Novel strategy for protein exploration:High-throughput screening assisted with fuzzy neural network[J].Journal of Molecular Biology 2005,351:683-692
[14]Coban R,Can B.Identification and control of ITU Triga Mark-Ⅱ Nuclear Research Reactor using neural networks and fuzzy logic[J].Advances in Artificial Intelligence,2005,3809:1057-1062
[15]Guo H J,Lin S F,Liu J H.A radial basis function sliding mode controller for chaotic Lorenz system[J].Physics Letters,2006,351:257-261
[16]Pai P F,Lin K P.Application of hybrid learning neural fuzzy systems in reliability predicton[J].Quality and Reliability Engineering International,2006,22:199-211
[17]Liu Y G,Li Y M.Sliding mode adaptive neural-network control for nonholonomie mobile modular manipulators[J].Journal of Intelligent & Robotic Systems,2005,44:203-224
[18]Yildirim S.Design of adaptive robot control system using recurrent neural network[J].Journal of Intelligent & Robotic Systems,2005,44:247-261
[19]Hansen L K,Salamon P.Neural network ensembles[J].IEEE Trans Pattern Analysis and Machine Intelligence,1990,12(10):993-1001
[20]Sollich P,Krogh A.Learning with ensembles:how over-fitting can be useful[J].Advances in Neural Information Processing Systems 8,Denver,CO:MIT Press,Cambridge,MA,1996:190-196
[21]Hansen L K,Liisberg L,Salamon P.Ensemble methods for handwritten digit recognition[J].In Proceedings of IEEE Workshop on Neural Networks for Signal Processing,Copenhagen,Denmark,1992:333-342
[22]Hampshire J,Waibel A.A novel objective function for improved phoneme recognition using timedelay neural networks[J].IEEE Transactions on Neural Networks,1990,1(2):216-228
[23]Gutta S,Wechsler H.Face recognition using hybrid classifier systems[J].In proceedings of IEEE International Joint Conference on Neural Network,Washington,1996:1017-1022
[24]Sharkey A J C,Sharkey N E,Cross S S.Adapting an ensemble approach for the diagnosis of breast cancer[J].In proceedings of International Conference on Artificial Neural Networks,Skovde,Sweden,1998:281-286
[25]Schapire R E,Singfer Y.BoosTexter:A boosting-based system for text categorization[J].Machine Learning,2000,39(2-3):135-168
[26]Dietterich T G.Ensemble methods in machine learning[J].In:Multiple Classifier Systems:Springer,2001:1-15
[27]周志华,陈世福.神经网络集成计算机学报[J].2002,25(1):1-8
[28]Valiant G.Atheory of the learnable[J].Communications of the ACM,1984,27(11):1134-1142
[29]Kearns M,Valiant G.Learning Boolean formulae or factoring[R].Technical Report TR-1488,Cambridge,MA:Harvard University Aiken Computation Laboratory,1988
[30]Krogh A,Vedelsby J.Neural network ensembles,cross validation,and active learning[J].In:Advances in Neural Information Processing Systems 7,CO:MIT Press,Cambridge,MA,1995:231-238
[31]傅强.选择性神经网络集成算法研究[D].浙江:浙江大学控制理论与工程,2007
[32]沈学华,周志华,吴建鑫,陈兆乾.Boosting和Bagging综述[J].计算机工程与应用,2000,12:31-33
[33]董乐红,耿国华,高原.Boosting算法综述[J].计算机应用与软件,2006,23(8):27-29
[34]Schapire R E.The strength of weak learnability[J].Machine Learning,1990,5(2):197-227
[35]Freund Y.Boosting a weak algorithm by majority[J].Information and Computation,1995,121(2):256-285
[36]Breiman L.Bagging predictors[J].Machine Learning,1996,24(2):123-140
[37]王正群,陈世福,陈兆乾.基于模糊划分的神经网络集成[J].南京大学学报(自然科学版),2006,42(1):63-67
[38]乐晓蓉,王正群,郭亚琴,王向东.基于聚类算法的神经网络集成[J].计算机工程,2007,33(5):180-184
[39]吕强,刘士荣,俞金寿.神经网络聚类加权集成及其应用[J].中南大学学报(自然科学版),2007,38(sup.l):32-35
[40]凌锦江,陈兆乾,周志华.基于特征选择的神经网络集成方法[J].复旦学报(自然科学版),2004,43(5):685-688
[41]凌锦江,周志华.基于因果发现的神经网络集成方法[J].软件学报,2004,15(10):1479-1484
[42]刘悦,吴耿锋,丁智国.基于特征加权的神经网络集成及其应用[J].上海大学学报(自然科学版),2007,13(5):623-628
[43]Lazarevic A,Kumar V.Feature bagging for outliner detection[J].KDD,Chicago,Illinois,USA,2005:157-166
[44]Partridge D,Yates W B.Engineering multiversion neural-net systems[J].Neural Computation,1996,8(4):869-893
[45]Maclin R,Shavlik J W.Combining the predictions of multiple classifiers:using competitive learning to initialize neural networks[J].Proceedings of the ~(14)th International Joint conference on Artificial Intelligence,Montreal,Canada,1995:524-530
[46]Optiz D W,Shavlik J W.Actively searching for an effective neural network ensemble[J].Connection Science,1996b,8(3-4):337-353
[47]巩文科,李长河,石争浩,赵洁.基于样本重构的神经网络集成学习方法[J].计算机应用,2006,26(6):1428-1430
[48]Islam M M,Yao X,Murase K.A constructive algorithm for training cooperative neural network ensembles[J].IEEE Transactions on Neural Netwoks,2003,14(4):820-834
[49]Liu Y,Yao X.Negatively correlated neural networks for classification[J].Proceedings of the 3~(rd)International Symposium on Artificial Life and Robotics,Beppu,Japan,1998:736-739
[50]Yong L,Xin Y.Simultaneous training of negatively correlated neural networks in ensemble systems[J].IEEE Transactions on Man and Cybernetics,Part B,1999,29(6):716-725
[51]Jang M,Cho S.Observational learning algorithm for an ensemble of neural networks[J].Pattern Analysis & Applications,2002,5(2):154-167
[52]王正群,陈世福,陈兆乾.并行学习神经网络集成方法.计算机学报[J].2005a,28(3):402-408
[53]王正群,陈世福,陈兆乾.一种主动学习神经网络集成方法.计算机研究与发展[J].2005b,42(3):375-380
[54]吴建鑫,周志华,沈学华,陈兆乾.一种选择性神经网络集成构造方法[J].计算机研究与发展,2000,37(9):1039-1044
[55]Zhihua Z,Jianxin W,Wei T.Ensembling neural networks:Many could be better than all[J].Artificial Intelligence,2002,137(122):239-263
[56]Lazarevic A,Obradoric Z.Effective pruning of neural network classifer ensembles[J].Processing International Joint conference on Neural Networks,2001:796-801
[57]Giacinto G,Roli F.Design of effective neural network ensembles for image classification purposes[J].Image and Vision Computing,2001,19:699-707
[58]Bakker B,Heskes T.Clustering ensembles of neural network models[J].Neural Networks,2003,16:261-269
[59]李凯,黄厚宽.一种基于聚类技术的选择性神经网络集成方法[J].计算机研究与发展,2005,42(4):594-598
[60]傅强,胡上序,赵胜颖.基于PSO算法的神经网络集成构造方法[J].浙江大学学报(工学版),2004,38(12):1596-1600
[61]凌霄汉,吉根林.一种快速选择性神经网络集成方法[J].郑州大学学报(理学版),2006,38(4):69-73
[62]施彦,黄聪明,侯朝桢.基于随机梯度法的选择性神经网络二次集成[J].计算机研究与发展2005,42(4):594-598
[63]傅向华,冯博琴,马兆丰,何明.增量构造负相关异构神经网络集成的方法[J].西安交通大学学报,2004,38(8):796-799
[64]陈国华.分层随机抽样方法与小子样检验[J].电子产品可靠性与环境试验,2001,5:11-13
[65]邓明.基于分层随机抽样的季节指数的抽样估计研究[J].统计研究,2008,25(7):70-75
[66]李洁,高新波,焦李成.一种基于修正模糊度的聚类有效性函数[J].系统工程与电子技术,2005,27(4):723-726
[67]诸克军,苏顺华,黎金玲.模糊c均值中的最优聚类和最佳聚类数[J].系统工程理论与实践,2005,25(3):52-61
[68]于剑,程乾生.模糊聚类方法中的最佳聚类数的搜索范围[J].中国科学(E辑),2002,32(2):274-280
[69]高新波,谢维信.模糊聚类理论发展及应用的研究进展[J].科学通报,1999,44(21):2241-2251
[70]Hecht-Nielson R.Theory of the back propagation neural networks[J].Neural Networks for perception,1992,2:65-93
[71]White H.Commentionist nonparametric regression:Multilayer feed forward networks can learn arbitrary mapping[J].Neural Networks,1990,3(4):535-549
[72]张振亚,王进,程红梅,王煦法.基于余弦相似度的文本空间索引方法研究[J].计算机科学,2005,32(9):160-163
[73]广凯,潘金贵.一种基于向量夹角的k近邻多标记文本分类算法[J].计算机科学,2008,35(4):205-207
[74]李希强,陆舍铭,王笛,刘春波,刘惠民,董学畅,缪明明.烟用香精UPLC指纹图谱的夹角余弦、相关系数与欧式距离评价[J].烟草科技,2008,12:37-41
[75]李俊峰,戴文战.基于夹角余弦的图像边缘检测算法及应用[J].中南大学学报(自然科学版),2007,38(sup.1):608-613
[76]C.L.Blake,C.J.Merz.UCI repository of machine learning databases[EB/OL].Http://www.ics.uci.edu/～mlearn/ML Repository.html.2007.12.12
[77]尚海燕.PTA残渣系统与溶剂系统建模[D].北京:北京化工大学控制理论与控制工程专业,2007