Research On Classification Algorithm And Optimization Of Support Vector Machine Based On Artificial Bee Colony

Support vector machines(SVM)have unique advantages in the classification of small sample data.The choice of parameters has an important impact on the classification accuracy and generalization ability of SVM.In view of the shortcomings of the current SVM parameter selection methods,an improved support vector machines optimization model based on artificial bee colony algorithm is proposed in this paper.Artificial bee colony(ABC)algorithm is an optimization algorithm proposed by Karaboga inspired by honey bee collecting behavior.As a swarm intelligence algorithm,the ABC algorithm searches the optimal solution of the optimization problem through the cooperation between different types of bees and the collection and sharing of food source information.As the solution search equation of ABC algorithm is good at exploration but poor at exploitation,the algorithm converges slowly,and it is easy to fall into local optimum.In this paper,an ABC algorithm with opposition-based learning(OLABC)is proposed.The OLABC algorithm firstly uses opposition-based learning to initialize the population to improve the quality of the initial solution.Secondly,to ensure the diversity of the population during the iteration of the algorithm,the idea of opposition-based learning is applied to the employed bee stage.When the quality of the newly generated solution is lower than that of the current solution,an opposite solution is generated and then use the greedy selection strategy to update solution,which further improves the global search ability of the algorithm.At the same time,in order to solve the problem of imbalance between the early exploration and the later exploitation capacity of the ABC algorithm,an adaptive weighting strategy is used to dynamically adjust the weight to balance the global and local search ability of the algorithm.Experiments on a set of benchmark test functions are conducted to verify the performance of the proposed algorithm.The experimental results show that compared with the traditional ABC algorithm and other classic improved algorithms,the ABC algorithm with opposition-based learning has better convergence speed and optimization accuracy.Based on the improvement of the ABC algorithm,take the parameters optimization of SVM as the problem to be optimized.By designing the fitness function,the improved ABC algorithm is used to optimize the penalty factor C and the kernel function parameter ?.Get the optimal parameter combination(C,?)of the SVM.The performance of the SVM classification model based on the improved ABC algorithm is verified on the UCI classic datasets.The results show that the proposed classification model has higher classification accuracy and better generalization ability.