EL-RFSVM: An Ensemble Learning Framework Based on Support Vector Machine and Random Forests for Labour Resource Allocation

The classical support vector machine (SVM) algorithm proposes only a two-class classification algorithm, but in the practical application of data mining, it is generally necessary to solve the classification problem of multiple classes. At the same time, for small data or low-dimensional data (data with few features), random forests do not produce good classifications. To address the limitations of single prediction models, this paper proposes a combined prediction model based on EL-RFSVM in the ensemble learning framework. This paper chooses a publicly available dataset of university students as the representative for labour allocation experiments. In this paper, we build two single prediction models, support vector machine model and random forest, and use the nonlinear mapping capability of the random forest to adjust the weight coefficients. The weights of the single model were determined to the two models, SVM and RF, and obtain the predicted labour allocation values. Finally, the performance of the three models on the problem is evaluated using AUC. The experimental results show that our proposed model has better prediction results compared to the single prediction model.