Improving Customer Value Index and Consumption Forecasts Using a Weighted RFM Model and Machine Learning Algorithms

Collecting and mining customer consumption data are crucial to assess value predict behaviors. This paper proposes a new procedure, based on an improved Random Forest Model by: adding indicator, joining the RFMS-based method K-means algorithm with Entropy Weight Method applied in computing index, classifying customers different categories, then constructing forecasting model whose RMSE is smallest all kinds of models. The results show that identifying by this RMF index facilitates profiling, enables development more precise marketing strategies.