The Research Of Simulation On Sorting Algorithm Based On Neural Network

Since the beginning of its application,BP neural network has been widely considered as an efficient network model.Meanwhile,many domestic and foreign scholars have taken abundant research about the approximation of linear functions and nonlinear functions.However,there islittle researchon algorithms that don't have clear mathematical relations at home and abroad by now.Therefore,a neural network based algorithm for sorting algorithm is proposed in this paper.Firstly,the BP neural network is used to make a prediction for the output of the sorting algorithm.Then,Hungarian algorithm and global greedy algorithm are used to balance the output and input data,which is the significant difference between sorting algorithm and most existing prediction models.The experimental results show that Hungarian algorithm and global greedy algorithm all have better effect on the sorting of data in low dimension,and the conclusion can be draw that the correct rate of the Hungarian algorithm is higher than the greedy algorithm in sorting according to a comprehensive comparison.The correct rate of the 10 dimensional data processed by Hungary algorithm and greedy algorithm is 75.70% and 44.85%,respectively.Take lower ranking accuracy for multi-dimensional data into consideration(the correct rate of 10 dimensional data sorting is less than90%),an optimization algorithm has been proposed by this paper.The accuracy in sorting of both algorithms have been improved to 77.70% and 76.35% for 10 dimensional data,but it still can't meet the requirements.The matching results are analyzed in detail,and the error compensation is introduced in the algorithm to improve the data of larger prediction error.Experimental results show that the method has a great improvement in the accuracy of sorting algorithm simulationand,the accuracy of 10 dimensional data sorting is 89.30% and 65.35% respectively after optimization.After Hungarian algorithm optimization,the sorting accuracy rate of 10 data reaches 90.26%,and the sorting accuracy rate for more dimensional data is also greatly improved.Finally,the sorting algorithm presented in this paper has a good simulation effect,which has achieved our research goal and will provide fundamental support in other algorithms simulation.