Construction And Empirical Study On User Incremental Operation Model Of Internet Financial Platform

Building a research model and discovering the laws behind it based on mining and analyzing user data,so as to provide decision support for realizing user incremental operation,is a new topic faced by internet financial platform.Limited by objective factors such as business model,user group and technology,there are many problems in the actual operation of Internet financial platform,such as unclear user positioning,inaccurate operation timing and mismatched operation strength.This paper will use data mining methods such as statistics and machine learning to build a research model of Internet financial platform user incremental operation and conduct empirical research,Realize the classification of credit and demand of Internet financial platform users,and then put forward the research on user value in combination with credit and demand,in order to solve the problem of incremental operation of Internet financial platform users.First,the paper briefs on the background and significance of the research.Then it reviews the related study on risk control of the Internet and user mining of Internet finance,and puts forward the objectives,contents and methods of the research.Second,it elaborates the concepts,theories and machine learning methods related to the research.The paper mainly expounds related concepts such as data mining and feature engineering.It introduces user research theories including RFM model,user segmentation theory,and AARRR user incremental operation.More importantly,it introduces machine learning algorithms,namely,logistic regression algorithm and K-means clustering algorithm.Based on the research results of RFM model,this paper puts forward the evolution from FAV-RFM model to FAVC-RFM model,points out the index system and design logic of FAVC-RFM model,and constructs the research model of user incremental operation of Internet financial platform.Based on the needs of building FAVC-RFM research model,this paper studies the credit rating and demand classification of users of Internet platform through data mining method,so as to provide basic analysis for the research of incremental operation model.Firstly,the credit rating of platform users is based on logistic regression method.The model effect is evaluated by accuracy rate,recall rate,AUC value and ROC value,and the users are graded from the perspective of credit rating.Secondly,the needs of platform users are classified based on K-means clustering algorithm,and the classification effect of K-means is evaluated by contour coefficient method and elbow method.Based on data mining,the results of credit classification and demand classification are obtained.Finally,it makes an empirical study on the user incremental operation model of Internet financial platform.Firstly,it analyzes the typical problems of user management of Internet financial platform from the dimensions of credit,demand and value.Then it makes an empirical study on the Internet user incremental operation analysis model based on improved FAVC-RFM.Finally,based on the idea of user classification,this paper puts forward corresponding measures and suggestions for each link of incremental operation,including the acquisition of channel users,the activation of sleeping users,the retention of post loan users,rapid and diversified realization and the recommendation of vertical fields,in order to achieve the goal of lower operation cost and higher user value.