Research On Genetic Algorithm In TSP Based On Hadoop

Information explosion era,the rapid increase in the size of data,Big data analysis and computing platform came into being.The intelligent algorithm,which is represented by genetic algorithm,has the characteristics of parallel operation,and it plays a more and more important role in dealing with many practical problems,such as multi condition,multi constraint and nonlinear.This thesis focuses on how to use genetic algorithms analysis run on the big data computing platform effectively,and instantating hybrid parallel genetic algorithm(HP G A,Hybrid Parallel Genetic Algorithm)proposed by this thesis in the TSP(traveling salesman problem-Travelling Salesman Problem).The main research work and achievements are as follows:(1)Discuss the possibility of parallel genetic algorithm.Because the genetic algorithm has the property of parallel operation.Therefore,how to improve the parallel effect is the focus of this thesis.In academic circles,many experts and scholars have done some research on the possibility of parallel genetic algorithm.On the basis of previous studies,this thesis makes a full discussion of the research results to ensure the optimization of the parallel characteristics of genetic algorithm.(2)To establish a hybrid parallel genetic algorithm model based on Hadoop.Based on the discussion of the feasibility of parallel genetic algorithm,a hybrid parallel genetic algorithm model based on Hadoop is constructed in this thesis.The parallel genetic algorithm model can be reasonably integrated to make it more flexible and improve the adaptability of the hybrid parallel genetic algorithm in solving the practical problems.At the same time,it can improve the efficiency of the genetic algorithm.(3)Instantiation and method of model framework.In this thesis,based on the parallel genetic algorithm,the TSP problem is used to verify the proposed model,and the experimental data are used to prove the availability of the model framework.