Analysis And Research Of Medical Device Sales Data Based On Data Mining

The sales process of the medical device industry has a profound impact on the medical device sales industry and the entire medical industry.It is of great significance to study the sales process data of the medical equipment and its sales industry and even the hospital.Medical device sales industry data mining is the discovery and use of its sales process data,to achieve deep-seated key technical means.This paper uses two kinds of data mining theories and methods to detect the sales process data and to forecast the sales volume in two aspects.The method of data mining of the sales data,and the actual data are validated separately.The main research work and the important research results are as follows:(1)Combining the existing anomaly detection method,the DBscan algorithm is used as the method of detecting the abnormal data of the sales data,and the K-means algorithm is used to select the adaptive parameters of the DBScan algorithm.The above method is used as the main method of the sales data anomaly detection model.(2)Combining with the existing data forecasting method,the stochastic forest algorithm based on Adaboost is used as the main method of sales forecasting.Then,by comparing the the random forest algorithm and the random forest algorithm based on Adaboost,the former data prediction method is selected as the main method of forecasting model.(3)By using the methods above respectively,with the actual sales process data on the forecast method and the anomaly detection method simulation experiments,the research proves the effectiveness of bothof the above methods.