Network Test System Based On Improved Genetic Algorithm Research And Realization

This article describes the design process of a network examination system. Through continuous study and improvement of the system integration of the Ant Colony Algorithm using the Genetic Algorithm as the algorithm for automatic test paper of network examinations.With the computer application technology and popularity of network-based, network examination from the study also tries to promote the use of continuous development. Network examination, compared with traditional forms of examination, in improving the efficiency of teachers, has a great advantage. The algorithm for automatic test paper is a core part of the network examination system what directly related to exam results. There are Common test paper method such as random test paper method, back test paper method and genetic algorithm. Simple random algorithm, defect is relatively large for too small or too large question bank; backtracking algorithm to solve the random algorithm in a larger problem:topic selection process stalled, the other with the same random algorithm, the current the most widely used. Genetic algorithms relatively better, but also more complicated to achieve, for a small test system to be too complex, currently at the research stage of actual use is relatively small.Genetic algorithm is a kind of reference of the biological laws of evolution (survival of the fittest, survival of the fittest) evolved out of an adaptive global optimization probability search method, which in a group all the individuals targeted, and use Randomized technical guidance on the parameters of an encoded search space efficiently. Genetic Algorithm in Algorithmic Research and applications have made some achievements. Ant colony (ant colony algorithm, referred to as ACA) is raised only in recent years a new type of simulated evolutionary algorithm. It is by the Italian scholars Dorigo, who received about the nature of collective behavior of real ant colony inspired research first put forward. Study found that when ant foraging ants are walking according to the most recent messenger path left by the amount of information to determine the best path.Genetic algorithm is fast global search capability, but the feedback system does not use often leads to inaction of the redundant iteration, solving low efficiency. Ant colony algorithm by improved genetic algorithm can be retained in the genetic processes of useful information, greatly accelerate the convergence of genetic algorithms. The system successfully used in the improved genetic algorithm as questions arithmetic, as do related research practical attempt.