Journal of Shanghai Jiaotong University ›› 2017, Vol. 51 ›› Issue (1): 98-.

Previous Articles     Next Articles

An Artificial Immune Network Classification Algorithm Based on VirtualForce Direction and Cell Differentiation

  

  1. Electronic Engineering Institute, Hefei 230037, China
  • Online:2017-01-31 Published:2017-01-31

Abstract:

Abstract: Aimed at the lack of effective methods to guide the evolution of antibodies in the traditional immune network classification algorithm, a virtual force and cell differentiation based artificial immune network classification algorithm (VCAINC) was proposed.The algorithm introduced the definition of virtual force zone to guide the  evolution of antibodies.Then, the moving convergence condition was introduced to determine the most suitable position of antibodies.For the antibodies which cannot satisfy the moving convergence condition, the strategy of cell differentiation was utilized to optimize the classifier. Finally, the performance of classification on UCI dataset was analyzed.The experimental results show that the VCAINC can effectively guide the evolution of antibodies and improve classification performance significantly.

Key words: artificial intelligence, artificial immune network, virtual force, cell differentiation

CLC Number: