Research On Financial Product Recommendation Based On Base Learning Algorithm Selection And Integrated Classification Of Immune Clonal Algorithm

With the advent of the era of big data and the development of financial technology,the number and types of financial products are becoming more and more,and more and more marketing activities have greatly weakened the public’s enthusiasm to buy.Moreover,with the increasingly fierce competition among financial institutions,major banks have to control the marketing efficiency and cost more intelligently and efficiently.Therefore,if we can purposefully recommend products and locate potential customers more accurately,we can not only greatly improve efficiency and reduce costs,but also avoid unnecessary interference to bank customers,resulting in resentment.It is a very important link for banks to maintain customers’ popularity in the process of marketing.The research method used in this paper is mainly based on immune clone feature selection and integrated classification to predict whether customers will buy regular financial products.The main contents and achievements can be summarized as the following two aspects.(1)An immune clonal algorithm is proposed to select the features of the data set,select multiple optimal feature subsets for classification,and make full use of the two sub classifier diversification methods of sample under sampling and feature selection,which further improves the classification performance.Artificial immune algorithm is a method to imitate the action principle of immune system in nature.(2)In this study,ensemble learning method is used for classification,and under sampling integration algorithm is used according to the characteristics of data set.These methods improve the classification performance and greatly improve the classification effect from the perspective of sample balance and diversity.Under sampling and feature selection ensure the randomness of the data,so the over fitting of the model can be reduced.