Parameter Optimization Of SVM Based On Multi-verse Optimizer

The performance of support vector machine has big relationship with parameter,but even now,there is not a theory to guide how to choose the optimal parameters.Then using the intelligent algorithm to solve this problem,Research contents:Aiming at Multi Verse Optimizer deficiency existing in exploitation,put forward two kinds of improvement methods.First increase the wormhole existence probability in every iteration for searching ability of algorithm,then constructing model and comparison with some classic SVM models.The results proved the validity of the improved model.Second to solve travel distance rate dropping slowly lead to its poor stability,offering some advice,and based on improved Multi Verse Optimizer proposed SVM model.The results indicate that performance of improved model is more stable and has strong generalization and learning ability.In order to validate the improved models,the two models are used for people with liver disease or not.The results show that the models are effective and feasible,and have good real applied perspective.