Application Of Adaboost-Based Ensemble Learning Method In Bank Marketing

With the rapid economic development,financial markets are also constantly innovating and changing.With the rapid development of machine learning and artificial intelligence,the distribution of commercial banks must also keep pace with the times,applying machine learning and artificial intelligence knowledge to continuously innovate and reform,and tap potential customers and potential value of customers.Based on the product marketing data of commercial banks,this paper uses Adaboost integrated learning algorithm combined with SVM and PCA algorithms to construct and implement Adaboost SVM algorithm.Multi-index evaluation is used in model evaluation,and it is compared with classic machine learning classification algorithms: SVM,decision tree and integration algorithms: random forest,xgboost,GBDT and simple neural network.It is found that the Adaboost SVM model has higher accuracy than the above models.Although the data has sample imbalance problems,the Adaboost SVM model also achieves a good fit,and there is no overfitting or underfitting problem.