Research And Improvement Of Multi-attribute Bilateral Matching Algorithm

With the popularization and development of information technology,when people deal with the problem of demand matching between supply and demand,the demand information that comes into contact is increasingly complicated and quantified.However,the existing bilateral matching model has low efficiency and complex model design process,which is difficult to meet the increasingly complex and variable matching requirements.Therefore,this thesis proposes an improved ant colony algorithm for multi-attribute bilateral matching problem and the solving model of dynamic multi-attribute bilateral matching problem.In solving multi-attribute bilateral matching problem by improved ant colony algorithm,this thesis abstractly modeled bilateral matching problem,improved attribute matching degree calculation model,and proposed non-linear gradient heuristic information and state transition strategy based on historical search information to solve the problems of early maturity and late convergence of ant colony algorithm.Aiming at the difficulty of setting initial parameters and adjusting parameters of ant colony algorithm,this thesis proposed a new method to solve the problem of multi-attribute bilateral match To solve the problem of heavy workload,an automatic parameter adjustment method based on gradient descent is proposed,and evaluation rules of stable matching and current optimal matching are formulated to guide pheromone updating of ant colony algorithm.In the solution model of dynamic multi-attribute bilateral matching problem,aiming at the problems existing in previous dynamic bilateral matching processing,such as the number of repeated matching times of individuals,the cost of solution,and the difficulty of dealing with the complex and changeable dynamic matching needs of massive matching individuals in ideal time,this thesis puts forward the concepts of influence function,estimation model of bilateral matching problem and re-matching set.The loudness function is used to measure the impact of demand changes on matching individuals in dynamic multi-attribute bilateral matching.The model of bilateral matching problem estimation is used to determine the final re-participation and individual set after demand changes,so as to reduce the number of individuals participating in matching and improve matching efficiency.In this thesis,simulation experiments are carried out on the model processing effect of improved ant colony algorithm for bilateral matching problem and dynamic multi-attribute bilateral matching problem.The experimental results show that the improved ant colony algorithm improves the solution effect and stability significantly compared with the traditional ant colony algorithm,and the improved ant colony algorithm based on RNA computing has better solution stability.The solving model of dynamic multi-attribute bilateral matching problem can significantly reduce the number of matching individuals participating in the matching,reduce the scale of the problem and improve the efficiency of solving the problem.