Classification of foreign language mobile learning strategy based on principal component analysis and support vector machine

Abstract

To improve the classification accuracy of foreign language mobile learning (m-learning) strategies applied by college students, an evaluation model based on principal component analysis (PCA) and support vector machine (SVM) is proposed. PCA was first employed to reduce the dimensionality of an evaluation system of foreign language m-learning strategies and the correlation between the indices in the system was eliminated. The first 5 principal components were extracted and a classification model based on SVM was established by taking the extracted principal components as its inputs. Gaussian radial basis function was adopted as the kernel function and the optimal SVM model was realized by adjusting the parameters C and g. The classification result was compared with those produced by a BP neural network model and a single SVM model. The simulation results prove that the PCASVM model has a simpler algorithm, faster calculating speed, higher classification accuracy and better generalization ability.