A General DNA Computing Model For Resolving The Classification Problem Of The Nerve Network

ANN(Artificial neural network)was the extremely popular interdisciplinary studies which recently developed. It involves discipline and so on biology, electron, computer, mathematics and physics which has the extremely widespread application background. The development of this discipline will have the important influence to present and the future science and technology development. Its unique structure and the method to deal with the information, caused it to obtain the remarkable result in many practical applications domain, and it could solve the problem which some traditional computers was extremely difficult to solve.DNA computing is an emerging new study area. What the subject studies is how to solve mathematical problems using DNA molecules. Since Adleman solve the Hamilton path problem using DNA molecules in 1994, many mathematical problems were solved by DNA computing such as matrix multiplication, addition, symbolic determinants. What is more, Denenson simulated the transformation of states of finite autonomous in Nov.2001. These facts demonstrate that DNA molecules computing by self-assembly according to Waston-Crick complement is powerful.In this paper, we established a general DNA computation model to solve the artificial neural network classification problem. The idea, which is different from traditional techniques that are used to modify powers between neurons, is to pick out the appropriate powers combination from any possible combination. The central feature of these models is to apply the parallel logic. Meanwhile by its general characteristic, this DNA computation model could be used for first-to-type network which has n level to solve its classified problem.