Design And Implementation Of O2O Goods Selecting System Based On Personalized Recommendation Algorithm

With the rapid development of the Internet,the e-commerce platform is also developing rapidly.In order to meet the needs of different consumer groups,the e-commerce platform continuously expands the number of goods and services.To improve efficiency of goods distribution and the offline stores sales,retail O2O(Online to Offline)services arises at the historic moment.Among many e-commerce enterprises who provide retail O2 O service,the low gross profit margin of the retail industry,how to rely on goods and services to stand out has become one of the most important thing of many retail O2 O service providers.There are two major differences between retail O2 O service and online sales.First of all,the difference between target users,retail O2 O services limited distribution range is limited by the shipping time,so target users is focus on store near area.While e-commerce platform online sales to the user in the entire network logistics distribution.Secondly,the quantity of goods,retail O2 O services limited by offline store warehouse space and cost,the quantity should be controlled within a certain range,while e-commerce platform is no limited of the goods quantity.There are few difference among traditional retail online sales,supermarkets,convenience stores and department stores,single channel solution cannot completely copy the application to the O2 O service.The resource of online and offline simply integrate cannot achieve the goal.If the competition between platforms is discount,that will eventually lead to the user's visiting decrease and user consumption decision-making influence over simple.In order to solve the problems faced by retail O2 O service,this paper focus on the O2 O service goods selecting model based on recommendation system,which provides retail O2 O service for different offline goods selecting solutions.By personalized recommendation service,consumers use less time on shopping.Consumers are more satisfied with the needed goods.Finally the O2 O service providers will become more competitive.The main work and innovation points are as follows:(1)Establish the user purchase preference model and goods selecting model.Building user preference model on the history of user purchasing record,and evaluation.Using goods selecting model helps O2 O service store to select goods for target user.(2)Design the goods selecting algorithm based on collaborative filtering algorithm.Design the selecting algorithm based on hybrid recommendation algorithm and the business needs.(3)Designed O2 O goods selecting system based on personalized recommendation algorithm and finished testing before the system was released.Goods selecting system was published on September 6th 2016.As of October 30,2016,test store sales amount was 23.12% above the general store sales amount.The user after purchasing rate is 17.83% above the general store.During the trial operation test the sales amount and the user after purchasing rate data show that the goods selecting system have a positive influence on the sales of O2 O store and customers.The algorithm is effective.