Research On The Recommendation Method Of Medical Platform

With the rapidly development of information technology,the medical sales platform has accumulated a large number of consumer data,and how to extract valuable information from it has become the focus of company.As of the end of 2019,more than 500 pharmaceutical trading companies in China have entered the field of e-commerce,and the competition in the pharmaceutical industry is becoming increasingly fierce.Therefore,companies urgently need to develop new ways to improve their competitiveness.At present,recommendation technology has been widely used in various industries.It can excavate consumer interest in massive data,recommend products that consumers like,improve consumer experience and merchant sales,and thus promote the development of merchants.China’s medical sales platform is mainly offline retailing,online e-commerce is mainly B2 B,B2C mode.This paper chiefly studies how to supply accurate recommendation services in the medical sales platform.For online medicine sales platform,we analyze a large amount of user consumption data,based on each user’s user portrait,using the current popular in-depth learning methods for accurate recommendation.For offline pharmaceutical sales platform,we use genetic algorithm to generate bundle medicals that can meet both the needs of consumers and the needs of merchants according to the law of consumers’ purchasing and the business needs of merchants.Then,we use collaborative filtering algorithm to predict the probability of consumers’ purchasing and personality preference factors,and build a mathematical model to get the optimal price of bundle medicals.The innovations of this article are as follows: 1.For online medicine platform,due to the large number of users and sparse scoring data,a recommendation technology based on deep neural network is proposed to predict the score of users who have not purchased medicines.2.For offline medical sales platform,due to the characteristics of unrecognized user identity and small coverage of users,a genetic algorithm is proposed to combine the purchase rules of users with the commercial needs of merchants,recommend drug combinations to offline medical sales platform,and price the bundle medicals.The theoretical analysis and experimental results demonstrate the validity of the proposed method,which has a high recommendation accuracy and meets the requirements of the medical platform.The pricing strategy model achieves an optimal balance between the purchase probability of the consumer and the total profit of the business among different constraints and conflicts.