基于ACC-RBF网络的脉象信号研究
|
摘要
现代脉象研究证实,脉象的形成,主要取决于心脏的功能、血管的机能、血液的质和量,脉象首先可显示这三方面因素的病变,其次可据其推断其他的病变。显然,脉象是有其客观存在因素的。是以确实可以通过分析脉象信号而达到辨明人体的病变。但是,脉象学多年来一直处于依赖语言表述和经验传授的阶段,有待进一步的深入研究、规范和统一,以促进其向客观、准确、科学、现代化发展。本论文在对人工神经网络(Artificial Neural Network, ANN)的基本概念、结构分类、训练函数和学习算法等基本理论进行详细阐述的基础上重点讨论了径向基函(Radial Basis Function, RBF)神经网络的相关性质,并使用RBF神经网络对22个正常人与22个海洛因成瘾者的脉象信号进行了分析和辨识。
径向基函数神经网络(RBF Neural Network, RBFNN)是一种三层前向型神经网络模型,具有良好的非线性全局逼近能力,而且结构简单,学习速度快,计算量少,目前已成功应用于模式识别、系统设计、函数逼近、信号处理、自适应滤波、非线性时间序列预测等方面。
蚁群聚类(Ant Colony Clustering, ACC)算法与粒子群优化(Particle Swarm Optimization, PSO)算法同属于群智能计算。本文针对ACC经典LF算法存在收敛速度慢的缺点,提出了一种改进的方法。使用这种方法设计了用于区别正常人与海洛因成瘾者脉象信号的RBF神经网络,利用粒子群优化算法对网络的权值进行优化。并且与使用k均值聚类法设计的网络进行了性能比较。最后对实验结果进行分析。
Modern research on the pulse signals indicates that the function of heart, the complexion of vas and the quality of blood response to pulse-feeling. So pulse signal can show health condition impersonality.
Radial basis function neural network (RBFNN) is a three-layer forward network。It have been extensively used in such diverse fields as pattern recognition, system design, function approach, signal processing, adaptive signal filter, nonlinear time series analysis and so on, owing to their features of simple architectures and brief training requirements.
In this paper we introduce an improved method of Ant Colony Clustering algorithm,then use the method to analyze the signal of pulse-feeling. Considering the characteristic differences between the pulse signals of heroin addicts and healthy persons, we successfully use RBF network based on K-means clustering and RBF network based on Ant Colony Clustering algorithm to identify heroin addicts from the pulse signals of 22 heroin addicts and 22 healthy persons.
The research shows that the network using ACC algorithm to identify the number of hidden nodes and PSO algorithm to train the weights of hidden layer has better performance of approximation and generalization than the network using K-means clustering algorithm with LMS algorithm.
径向基函数神经网络(RBF Neural Network, RBFNN)是一种三层前向型神经网络模型,具有良好的非线性全局逼近能力,而且结构简单,学习速度快,计算量少,目前已成功应用于模式识别、系统设计、函数逼近、信号处理、自适应滤波、非线性时间序列预测等方面。
蚁群聚类(Ant Colony Clustering, ACC)算法与粒子群优化(Particle Swarm Optimization, PSO)算法同属于群智能计算。本文针对ACC经典LF算法存在收敛速度慢的缺点,提出了一种改进的方法。使用这种方法设计了用于区别正常人与海洛因成瘾者脉象信号的RBF神经网络,利用粒子群优化算法对网络的权值进行优化。并且与使用k均值聚类法设计的网络进行了性能比较。最后对实验结果进行分析。
Modern research on the pulse signals indicates that the function of heart, the complexion of vas and the quality of blood response to pulse-feeling. So pulse signal can show health condition impersonality.
Radial basis function neural network (RBFNN) is a three-layer forward network。It have been extensively used in such diverse fields as pattern recognition, system design, function approach, signal processing, adaptive signal filter, nonlinear time series analysis and so on, owing to their features of simple architectures and brief training requirements.
In this paper we introduce an improved method of Ant Colony Clustering algorithm,then use the method to analyze the signal of pulse-feeling. Considering the characteristic differences between the pulse signals of heroin addicts and healthy persons, we successfully use RBF network based on K-means clustering and RBF network based on Ant Colony Clustering algorithm to identify heroin addicts from the pulse signals of 22 heroin addicts and 22 healthy persons.
The research shows that the network using ACC algorithm to identify the number of hidden nodes and PSO algorithm to train the weights of hidden layer has better performance of approximation and generalization than the network using K-means clustering algorithm with LMS algorithm.
引文
[1]蔡自兴等.人工智能及其应用(第3版) [M].北京:清华大学出版社, 2003.
[2]韦莉萍,周珉.海洛因依赖者肝炎病毒及HIV感染情况研究进展[J].解放军预防医学杂志,1999,Vol.17,No.5.
[3] Haykin.神经网络原理[M].机械工业出版社,2004.
[4]李冰星,朱文锋,黄惠勇.中医脉诊研究思考[J].中国中医基础医学杂志,1998.Vol.4, No.6.20-22.
[5]张力.脉诊研究进展.山东中医学院学报[J].1995,Vol.19, No.4:281-284.
[6]蒋宗礼.人工神经网络导论[M].高等教育出版社,2001.
[7]陈益,陈嘉琳,连文.228例吸毒者的肝功能和乙肝感染指标调查[J].广东卫生防疫.1995.Vol.21, No.3.
[8] F. Rosenblatt. The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain[J]. Psychological Review,vol.65,1958, pp.386-408. Reprinted in 1988.Anderson and Rosenfeld[1].pp.92-114.
[9]杨福生,高上凯.生物医学信号处理[M].高等教育出版社,1989.
[10] Hagan M T. Neural Network Design[M],机械工业出版社.2002.
[11]袁曾任.人工神经网络及应用[M].北京.清华大学出版社.1999.
[12]张兵,李唯利等.神经网络理论的发展与前沿问题[J].湖南电力.2001年第4期
[13]王璐,吴南健,温殿中.人工神经网络在孕妇脉象判别中的应用[J].黑龙江大学自然科学学报. Vol.20,No.4,2003.
[14] Gorisi F , Poggio T. Networks and the best approximation property[J]. Biological Cybernetics,1990(63):169~176.
[15] Bornhold A. General Asymmetric Neural Networks and Structure Design by Genetic Algorithms[J]. Neural Networks,1992, 5(2): 327-334.
[16] David J J, Frenzel J F. Trainining Product Unit Neural Networks with Genetic Algorithms[C].IEEE Expert, 1993, 8(5): 26-33.
[17]李春宇.基于RBF神经网络的中医脉象信号检测[C].重庆大学硕士学位论文,2006.
[18]姜涛.基于小波分析的中医脉象识别研究[C].河海大学硕士学位论文,2003.
[19] Powell M J D. Radial Basis Function for multivariable interplation[C], Review IMA Conf. on Algorithms for the Approximation of Functions and Data. RMCS Shrivanham, 1985.
[20] D. S. Broomhead, D. Lowe. Multivariable function interpolation and adaptive network[J]. Complex Systems, 1988(2).
[21] J. Moody, C. Darken. Fast Learning in Networks of Locally-tuned Processing units[J]. Neural Computation, 1989(1):281~294.
[22] Broomhead D S, et al. Multivariable function interpolation and adaptive networks[J]. Complex Systems, 1988,2:321~355.
[23] Mulgrew Bernard. Applying radial basis functions[J]. IEEE Signal Processing Magazine,1996,13(2):50~65.
[24] Miccheli C. Interpolation of scattered data: distance matrices and conditionally positive definite functions[J]. Constructive Approximation, 1986(2): 11~22.
[25] Moody J E. The effective number of parameters: an analysis of generalization and regularization in nonlinear learning systems [A]. In Advances in Neural Information Processing Systems4[C], San Mateo, CA,1992,847-85.
[26]高隽.人工神经网络原理及仿真实例[C].机械工业出版社.2003年8月第1版.
[27] Simon Haykin.神经网络的综合基础(第2版)(影印版)[M].清华大学出版社,2001.11.
[28] Tomason Poggio, Federico Girosi. Networks for approximation and learning[J]. Proceedings of the IEEE.1990, Vol.78, No.9.
[29]钟志强.基于径向基神经网络和遗传算法建立转炉提钒终点预报模型的研究[C].重庆大学硕士学位论文,2002.
[30] Friedhelm Schwenker, Hans A. Kestler, GuEnther Palm, Three learning phases for radial-basis-function-networks[J]. Neural Networks, 2001 (14): 439~458.
[31] Simon Haykin. Neural networks: A Comprehensive Foundation[M]. Upper Saddle River, NJ: Prentice Hall, 1999.
[32] Chen S, Cowan C F N and Grant P M. Orthogonal least squares learning algorithms for radial basis function networks [J]. IEEE Trans. Neural Networks, 1991,2(2):302-309.
[33] Sherstinsky A , Picard R W. On the efficiency of the orthogonal least squares training method for radial basis function networks [J]. IEEE Trans. Neural Networks, 1996 , 7(1):195-200.
[34] Moody J And Darken C. Fast Learning in Networks of Locally-Tuned Processing Units[J]. Neural Computation,1989(1): 281~294.
[35] Platt J. A resource-allocating network for function interpolation[J] ,1991(3):213~225.
[36] Minsky M. AI Magazine[M], Summer, 34-51, 1991.
[37] Bonabeau E, Dorigo M, Theraulaz G. Swarm Intelligence: From Natural to Artificial Systems [M],OxfordUniv. Press, New York,1999.
[38]吴斌.群体智能的研究及其在知识发现中的应用[C].中国科学院研究生院博士论文.2002.
[39] M. Dorigo. Optimiztion, Learning and Natural Algorithma(in Italian)[D]. Ph.D. thesis,Dipartimento di Elettronica, Politecnico di Milano, IT, 1992.
[40] A. Colorni, M. Dorigo, V. Maniezzo. Distributed optimization by ant colonies[C]. InProceedings of the First European Conference onArtificial Life, pages 134-142 Elsevier, 1992.
[41] L. M. Gambardella , M. Dorigo. Ant-Q: A reinforcement learning approach to the travelingsalesman problem[C]. In A. Prieditis and S. Russell, editors, Proceedings of the TwelfthInternational Conference on Machine Learning (ML-95), pages 252–260. Morgan KaufmannPublishers, PaloAlto, CA, 1995.
[42] T. Stützle , H. Hoos. The MAX–MIN ant system and local search for the traveling salesmanproblem[C]. In T. Baeck, Z. Michalewicz, and X. Yao, editors, Proceedings ofIEEE-ICEC-EPS’97, IEEE International Conference on Evolutionary Computation and Evolutionary Programming Conference, pages 309–314. IEEE Press, 1997.
[43] T. Stützle , H. Hoos. Improvements on the ant system: Introducing MAX–MIN ant system[C]. In Proceedings of the International Conference on Artificial Neural Networks and GeneticAlgorithms, pages 245–249. Springer Verlag, Wien, 1997.
[44] T. Stützle and H. Hoos. MAX–MIN Ant system and local search for combinatorial optimization problems[M]. In S. Vo?, S. Martello, I.H. Osman, and C. Roucairol, editors,Meta-Heuristics: Advances and Trends in Local Search Paradigms for Optimization, pages137–154. Kluwer, Boston, 1998.
[45] T. Stützle MAX-MIN Ant System for Quadratic Assignment Problems[R]. Technical ReportAIDA-97-04, Intellectics Group, Department of Computer Science, Darmstadt University of Technology, Germany, July 1997.
[46] M. Dorigo and L. M. Gambardella. Ant colony system: A cooperative learning approach to the traveling salesman problem[J]. IEEE Transactions on Evolutionary Computation, 1(1):53–66,1997.
[47]韩家炜著范明等译数据挖掘概念与技术[M],机械工业出版社,2001,
[48] J Han, M Kamber. Data Mining: Concepts and Techniques[M].北京:高等教育出版社, 2001.
[49]冯永.基于计算智能的聚类技术及其应用研究[C].重庆大学博士论文.2004.
[50]杨新斌,孙京诰,黄道.一种进化聚类学习新方法[J].计算机工程与应用, 2003; 39(15) : 60～62.
[51] Bonabeau E, Dorigo M, Theraulaz G. Swarm Intelligence: From Natural to Artificial Systems [M],Oxford Univ. Press, New York,1999.
[52] Deneubourg J L, Goss S, Franks N, et al. The dynamics of collective sorting: robot-like ants and ant-like robots[C]. Proceedings of the 1st International Conference on Simulation ofAdaptive Behavior: From Animals to Animals, 1991,356~363.
[53] Lumer E, Faieta B. Diversity and adaptation in populations of clustering ants . Proceedings of the Third International Conference on Simulation of Adaptive Behavior: From Animals to Animats[J], Vol.3, MIT Press/ Bradford Books, Cambridge, MA, 1994, pp 501-508.
[54] N Monmarche, M Slimane, G Venturiui. On Improving Clustering in Numerical Databases with Artificial Ants 5th. [M] European Conference on Artificial life (ECAL'99). Lecture Notes in Artificial intelligence, 1999, 1674: 626-635.
[55] N Monmarche, M Slimane, G Venturiui. AntClass: Discovery of Clusters in Numeric Data by an Hybridization of Ant Colony with the Kmeans Algorithm[C]. http://citeseer.nj.nec.com.
[56] N Monmarche. On Data Clustering with Artificial Ants, Data Mining with Evolutionary Algorithms: Research Directions[C]. AAAI Press,1999: 23-26.
[57]郝晋,石立宝,周家启.具有随机扰动的蚁群算法[J].系统工程理论与实践,1999,3:84~87.
[58]李万庆,李彦苍.求解复杂优化问题的基于信息熵的自适应蚁群算法[J].数学的实践与认识.2005,35(2):134～139.
[59]段海滨.蚁群算法原理及其应用[M].科学出版社,2005.
[60]王万良,唐宇.微粒群算法的研究现状与展望[J].浙江工业大学学报,2007,35(2):136-141.
[61] Schafer J D,Whitley D.Proceedings of the Workshop on Combinations of GeneticAlgorithms and Neural Networks[M], 1992,Los Alarnitos,CA:The IEEE Computer SocietyPress,1992.
[2]韦莉萍,周珉.海洛因依赖者肝炎病毒及HIV感染情况研究进展[J].解放军预防医学杂志,1999,Vol.17,No.5.
[3] Haykin.神经网络原理[M].机械工业出版社,2004.
[4]李冰星,朱文锋,黄惠勇.中医脉诊研究思考[J].中国中医基础医学杂志,1998.Vol.4, No.6.20-22.
[5]张力.脉诊研究进展.山东中医学院学报[J].1995,Vol.19, No.4:281-284.
[6]蒋宗礼.人工神经网络导论[M].高等教育出版社,2001.
[7]陈益,陈嘉琳,连文.228例吸毒者的肝功能和乙肝感染指标调查[J].广东卫生防疫.1995.Vol.21, No.3.
[8] F. Rosenblatt. The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain[J]. Psychological Review,vol.65,1958, pp.386-408. Reprinted in 1988.Anderson and Rosenfeld[1].pp.92-114.
[9]杨福生,高上凯.生物医学信号处理[M].高等教育出版社,1989.
[10] Hagan M T. Neural Network Design[M],机械工业出版社.2002.
[11]袁曾任.人工神经网络及应用[M].北京.清华大学出版社.1999.
[12]张兵,李唯利等.神经网络理论的发展与前沿问题[J].湖南电力.2001年第4期
[13]王璐,吴南健,温殿中.人工神经网络在孕妇脉象判别中的应用[J].黑龙江大学自然科学学报. Vol.20,No.4,2003.
[14] Gorisi F , Poggio T. Networks and the best approximation property[J]. Biological Cybernetics,1990(63):169~176.
[15] Bornhold A. General Asymmetric Neural Networks and Structure Design by Genetic Algorithms[J]. Neural Networks,1992, 5(2): 327-334.
[16] David J J, Frenzel J F. Trainining Product Unit Neural Networks with Genetic Algorithms[C].IEEE Expert, 1993, 8(5): 26-33.
[17]李春宇.基于RBF神经网络的中医脉象信号检测[C].重庆大学硕士学位论文,2006.
[18]姜涛.基于小波分析的中医脉象识别研究[C].河海大学硕士学位论文,2003.
[19] Powell M J D. Radial Basis Function for multivariable interplation[C], Review IMA Conf. on Algorithms for the Approximation of Functions and Data. RMCS Shrivanham, 1985.
[20] D. S. Broomhead, D. Lowe. Multivariable function interpolation and adaptive network[J]. Complex Systems, 1988(2).
[21] J. Moody, C. Darken. Fast Learning in Networks of Locally-tuned Processing units[J]. Neural Computation, 1989(1):281~294.
[22] Broomhead D S, et al. Multivariable function interpolation and adaptive networks[J]. Complex Systems, 1988,2:321~355.
[23] Mulgrew Bernard. Applying radial basis functions[J]. IEEE Signal Processing Magazine,1996,13(2):50~65.
[24] Miccheli C. Interpolation of scattered data: distance matrices and conditionally positive definite functions[J]. Constructive Approximation, 1986(2): 11~22.
[25] Moody J E. The effective number of parameters: an analysis of generalization and regularization in nonlinear learning systems [A]. In Advances in Neural Information Processing Systems4[C], San Mateo, CA,1992,847-85.
[26]高隽.人工神经网络原理及仿真实例[C].机械工业出版社.2003年8月第1版.
[27] Simon Haykin.神经网络的综合基础(第2版)(影印版)[M].清华大学出版社,2001.11.
[28] Tomason Poggio, Federico Girosi. Networks for approximation and learning[J]. Proceedings of the IEEE.1990, Vol.78, No.9.
[29]钟志强.基于径向基神经网络和遗传算法建立转炉提钒终点预报模型的研究[C].重庆大学硕士学位论文,2002.
[30] Friedhelm Schwenker, Hans A. Kestler, GuEnther Palm, Three learning phases for radial-basis-function-networks[J]. Neural Networks, 2001 (14): 439~458.
[31] Simon Haykin. Neural networks: A Comprehensive Foundation[M]. Upper Saddle River, NJ: Prentice Hall, 1999.
[32] Chen S, Cowan C F N and Grant P M. Orthogonal least squares learning algorithms for radial basis function networks [J]. IEEE Trans. Neural Networks, 1991,2(2):302-309.
[33] Sherstinsky A , Picard R W. On the efficiency of the orthogonal least squares training method for radial basis function networks [J]. IEEE Trans. Neural Networks, 1996 , 7(1):195-200.
[34] Moody J And Darken C. Fast Learning in Networks of Locally-Tuned Processing Units[J]. Neural Computation,1989(1): 281~294.
[35] Platt J. A resource-allocating network for function interpolation[J] ,1991(3):213~225.
[36] Minsky M. AI Magazine[M], Summer, 34-51, 1991.
[37] Bonabeau E, Dorigo M, Theraulaz G. Swarm Intelligence: From Natural to Artificial Systems [M],OxfordUniv. Press, New York,1999.
[38]吴斌.群体智能的研究及其在知识发现中的应用[C].中国科学院研究生院博士论文.2002.
[39] M. Dorigo. Optimiztion, Learning and Natural Algorithma(in Italian)[D]. Ph.D. thesis,Dipartimento di Elettronica, Politecnico di Milano, IT, 1992.
[40] A. Colorni, M. Dorigo, V. Maniezzo. Distributed optimization by ant colonies[C]. InProceedings of the First European Conference onArtificial Life, pages 134-142 Elsevier, 1992.
[41] L. M. Gambardella , M. Dorigo. Ant-Q: A reinforcement learning approach to the travelingsalesman problem[C]. In A. Prieditis and S. Russell, editors, Proceedings of the TwelfthInternational Conference on Machine Learning (ML-95), pages 252–260. Morgan KaufmannPublishers, PaloAlto, CA, 1995.
[42] T. Stützle , H. Hoos. The MAX–MIN ant system and local search for the traveling salesmanproblem[C]. In T. Baeck, Z. Michalewicz, and X. Yao, editors, Proceedings ofIEEE-ICEC-EPS’97, IEEE International Conference on Evolutionary Computation and Evolutionary Programming Conference, pages 309–314. IEEE Press, 1997.
[43] T. Stützle , H. Hoos. Improvements on the ant system: Introducing MAX–MIN ant system[C]. In Proceedings of the International Conference on Artificial Neural Networks and GeneticAlgorithms, pages 245–249. Springer Verlag, Wien, 1997.
[44] T. Stützle and H. Hoos. MAX–MIN Ant system and local search for combinatorial optimization problems[M]. In S. Vo?, S. Martello, I.H. Osman, and C. Roucairol, editors,Meta-Heuristics: Advances and Trends in Local Search Paradigms for Optimization, pages137–154. Kluwer, Boston, 1998.
[45] T. Stützle MAX-MIN Ant System for Quadratic Assignment Problems[R]. Technical ReportAIDA-97-04, Intellectics Group, Department of Computer Science, Darmstadt University of Technology, Germany, July 1997.
[46] M. Dorigo and L. M. Gambardella. Ant colony system: A cooperative learning approach to the traveling salesman problem[J]. IEEE Transactions on Evolutionary Computation, 1(1):53–66,1997.
[47]韩家炜著范明等译数据挖掘概念与技术[M],机械工业出版社,2001,
[48] J Han, M Kamber. Data Mining: Concepts and Techniques[M].北京:高等教育出版社, 2001.
[49]冯永.基于计算智能的聚类技术及其应用研究[C].重庆大学博士论文.2004.
[50]杨新斌,孙京诰,黄道.一种进化聚类学习新方法[J].计算机工程与应用, 2003; 39(15) : 60～62.
[51] Bonabeau E, Dorigo M, Theraulaz G. Swarm Intelligence: From Natural to Artificial Systems [M],Oxford Univ. Press, New York,1999.
[52] Deneubourg J L, Goss S, Franks N, et al. The dynamics of collective sorting: robot-like ants and ant-like robots[C]. Proceedings of the 1st International Conference on Simulation ofAdaptive Behavior: From Animals to Animals, 1991,356~363.
[53] Lumer E, Faieta B. Diversity and adaptation in populations of clustering ants . Proceedings of the Third International Conference on Simulation of Adaptive Behavior: From Animals to Animats[J], Vol.3, MIT Press/ Bradford Books, Cambridge, MA, 1994, pp 501-508.
[54] N Monmarche, M Slimane, G Venturiui. On Improving Clustering in Numerical Databases with Artificial Ants 5th. [M] European Conference on Artificial life (ECAL'99). Lecture Notes in Artificial intelligence, 1999, 1674: 626-635.
[55] N Monmarche, M Slimane, G Venturiui. AntClass: Discovery of Clusters in Numeric Data by an Hybridization of Ant Colony with the Kmeans Algorithm[C]. http://citeseer.nj.nec.com.
[56] N Monmarche. On Data Clustering with Artificial Ants, Data Mining with Evolutionary Algorithms: Research Directions[C]. AAAI Press,1999: 23-26.
[57]郝晋,石立宝,周家启.具有随机扰动的蚁群算法[J].系统工程理论与实践,1999,3:84~87.
[58]李万庆,李彦苍.求解复杂优化问题的基于信息熵的自适应蚁群算法[J].数学的实践与认识.2005,35(2):134～139.
[59]段海滨.蚁群算法原理及其应用[M].科学出版社,2005.
[60]王万良,唐宇.微粒群算法的研究现状与展望[J].浙江工业大学学报,2007,35(2):136-141.
[61] Schafer J D,Whitley D.Proceedings of the Workshop on Combinations of GeneticAlgorithms and Neural Networks[M], 1992,Los Alarnitos,CA:The IEEE Computer SocietyPress,1992.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |