An Ensemble Method for Medicine Best Selling Prediction

摘要

In this paper the author presents a data mining model based ensemble methodology in medicine selling prediction. The ANN algorithm is utilized for feature selection, and some machine learning techniques are constructed for ensemble model in medicine best selling and best sell season prediction. The verification is conducted on the real cases from GuiYang ZhiFuTang Pharmacy Company Sell data which consists of 63044 records and 17 fields. The real case experiments show that, our approach works effectively and could be used as an assistant approach for selling analysis in some circumstances.