Value Forcast Of Railway Freight Based On Data Mining Technology

With the improvement of transportation technology,the railway freight market has become increasingly prosperous.The railway freight market is facing increasingly fierce competition simultaneously because the development of e-commerce platforms has raised the requirements of the market on railway freight efficiency and service capacity.It is an important part of CRM research to grasp the historical behavior data of freight customers and predict their future value in advance.This paper takes the value prediction of railway freight customers as the target and data mining technology as the support,analyzes the relevant data of railway freight customers,introduces the concept of data mart,and designs a data mart for the value prediction of railway freight customers used by the railway business department.Based on the traditional RFM model,the evaluation index of railway freight customer value suitable for railway freight industry is selected,and the improved evaluation method of RFM railway freight customer value is proposed.Based on the characteristics of railway freight customers,the random forest algorithm model is selected to train the railway freight customer value prediction classifier,and the railway freight value prediction model is constructed.An example is given to validate and evaluate the model of railway freight customer value prediction.Finally,the application of railway freight customer value prediction model in railway freight customer portrait and early warning of railway freight customer value change is analyzed.The results show that: The railway freight customer value prediction model based on data mining has a relatively good ability to predict the railway freight customer value.And railway freight customer value prediction has a wide range of application value.