Research On Classification Of International Exchange And Cooperation Level Of Colleges And Universities Based On Support Vector Machine

International exchange and cooperation is an important way to the development of colleges and universities, therefore,the research combining qualitative and quantitative analysis has a great social significance and practical significance as well.Considering its data small sample and nolinear characteristics,this research uses support vector machine which has classification accuracy and generalization ability to create two sets of support vector machine model for the international exchange and cooperation based on bionic intelligent algorithm.The GA_SJ algorithm is proposed.It is a model of support vector machine classification to obtain the automatic kernel parameter selection and optimization for SVM training.With introducing the random search,the optimal preservation strategy,dynamic crossover and mutation probability into the genetic algorithm,this algorithm effectively improves the efficiency of the genetic algorithm,and increases the classification ability of the quantitative data on international exchange and cooperation in colleges and universities.It creates the ACO-SVM model.This model uses the classification and prediction accuracy rate of the SVM as the objective function and improves the ant colony algor-ithmhm.It also introduces directed search and pheromone updating rule,which is based on the time-varying function to update information.By employing the parallelism,positi ve feedback and robustness of ant colony algorithm.this model combines SVM to obtain the optimal target and the optimal parameter combination.It refines the classification a nalysis of the quantitative data on international exchange and cooperation in colleges an d universities.This research creates two sets of positioning frameworks for international exchange and cooperation in colleges and universities.These two frameworks are based on the genetic algorithm and random thoughts phase coupling algorithm,and they advance the ant colony algorithm and optimize the parameters of SVM model construction.This study is used for the positioning of international exchange and cooperation level of colleges and universities in Liaoning,and has achieved good predicted classification effect.As a result,this study provides two feasible qualitative methods to analyze the positioning of international exchange and cooperation level in colleges and universities.