Neural Network Methods for Construction of Sociodynamic Models Hierarchy

详细信息    查看全文

• 关键词：Sociodynamics ; Artificial neural networks ; Markov processes ; Kolmogorov equation ; Fokker ; Planck equation ; Hierarchical systems ; Multiagent systems
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9719
• 期：1
• 页码：513-520
• 全文大小：185 KB
• 参考文献：1.FuturICT. http://​www.​futurict.​eu/​
  2.Putin’s speech at the plenary session of the 16th St. Petersburg international economic forum. http://​kremlin.​ru/​transcripts/​15709
  3.Tarkhov, D.A., Vasilyev, A.N.: Mathematical models of complex systems on the basis of artificial neural networks. Nonlin. Phenom. Complex Syst. 17(3), 327–335 (2014)MATH
  4.Kainov, N.U., Tarkhov, D.A., Shemyakina, T.A.: Application of neural network modeling to identification and prediction in ecology data analysis for metallurgy and welding industry. Nonlin. Phenom. Complex Syst. 17(1), 57–63 (2014)MATH
  5.Weidlich, W.: Sociodynamics: A Systematic Approach to Mathematical Modelling in the Social Sciences. Harwood Academic, Amsterdam (2000)MATH
  6.Helbling, D.: Quantative Sociodynamics. Springer, Heidelberg, Dordrecht, London, New York (2010)CrossRef
  7.Polyanin, A.D.: Spravochnik po linejnym uravneniyam matematicheskoj fiziki. FizMatLit, Moscow (2001)
  8.Mosalov, O.P., Red’ko, V.G., Prokhorov, D.V.: Simulation of the evolution of autonomous adaptive agents. Math. Models Comput. Simul. 1(1), 156–164 (2009)CrossRef MATH
  
• 作者单位：Ekaterina A. Blagoveshchenskaya (16) (17)
  Aleksandra I. Dashkina (17)
  Tatiana V. Lazovskaya (17)
  Viktoria V. Ryabukhina (16)
  Dmitriy A. Tarkhov (17)
  
  16. Petersburg State Transport University, 9 Moskovsky pr., 190031, Saint Petersburg, Russia
  17. Peter the Great St. Petersburg Polytechnical University, 29 Politechnicheskaya Street, 195251, Saint Petersburg, Russia
  
• 丛书名：Advances in Neural Networks ¨C ISNN 2016
• ISBN：978-3-319-40663-3
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：9719

文摘

The article includes the following modern approaches to modelling sociodynamic processes: Kolmogorov equation system for Markov process with a discrete set of states, Fokker-Planck equation, multiagent systems, etc. As an example, one demographic task of predicting is solved. We compare the simplest neural network approach with an approach based on a special evolutionary model. The article also justifies applying the neural network modelling for producing solutions to the above mentioned equations, determination of their coefficients on the basis of observations and making more precise models, including the dependence of human behaviour on a psychological type. A possibility of making models more precise as new data come in has been discussed.