The Mechanism And Significance Of Genetic Algorithm Pattern Recognition

Based on the artificial intelligence in the field of science to select the typical genetic algorithm as the research object,to deeply discussed the mechanism of genetic algorithm and pattern recognition,control,operator,algorithm adaptive mechanism of this a few aspects of the detail,reflect the essential connotation of human science,based on the understanding of the mechanism of this algorithm to realize algorithm is the digital model to the real problem,into computer language logic thinking method to the human world.The exploration of the nature,development process and function of scientific theory provides a new theoretical basis and practical development direction for the development of philosophy of science and technology.In this paper,the principles of genetic algorithm and pattern recognition are briefly summarized.As a representative of the core algorithm in the field of artificial intelligence,genetic algorithm has a good regulation strategy and parallel search performance,which can well reflect the intelligent characteristics of the development of the field of artificial intelligence.Therefore,genetic algorithm is taken as the main research object of pattern recognition.The operation mechanism of pattern recognition of genetic algorithm is mainly reflected in the adjustment mechanism of genetic operator,the setting mechanism of adaptive function and parallelism.Firstly,it can be concluded from the research on the adjustment mechanism that a good screening mechanism of the selection operator can preserve the individual advantages of the sample,expand the excellent performance of the population and enhance the evolutionary process of the population.The characteristic of crossover operator is to provide help for expanding the diversity of population,so as to find new gene space and make the individuals in the new population contain diversity.Mutation operator is to make changes to gene points at certain gene positions of individual strings in the target population.The purpose is also to maintain the diversity of the population,generate a dynamic balance with crossover operator in terms of functionality,and reflect the adaptive characteristics of the algorithm.But the real balance need to rely on the fitness function,to adapt to the function of setting can directly affect the results of the algorithm,if the algorithm using adaptive value difference between high selection strategy,so will lead to the advantage of individual algorithm premature convergence to local optimal solution,on the contrary,if the individual to adapt to the value of the difference is too small,so can lead to the slow convergence speed of the algorithm and stagnation phenomenon.Therefore,the existence of the adjustment operator reflects the adaptive performance of genetic algorithm pattern recognition,which needs to be adjusted according to the change of external conditions.Secondly,Second,from the parallel characteristic,this is the schema theorem itself has the characteristics of genetic algorithm,because the genetic algorithm is based on population combinations group search,a small scale and search the solution space in many regions,mutual exchange of information,to evaluate multiple solution of search space,make the algorithm has good global search capability,and according to the principle of the schema theorem for fitness is higher than the mean fitness,length is shorter,the low-order model classes in the iteration of genetic algorithm according to the index law of growth,this suggests that the genetic algorithm of the cause of the evolution,The effective models that can survive in the process of genetic calculation are all the excellent models with short length,low order and average adaptive value higher than the average adaptive value of the population.Here,Occam's razor theory is applied and simple results in high efficiency.Finally,through the mechanism research of pattern recognition of genetic algorithm,the essence of machine cognition is profoundly reflected.It is realized that the algorithm reflects the mechanical intelligence of computer,and the digital modeling of real problems is transformed into the thinking method of computer logic language to understand the human world.Since it is a mechanical calculation step,there will be a boundary problem that cannot be calculated,so the boundary problem of machine cognition is an essential philosophical exploration for the future development of artificial intelligence society.