Research On Supermarket Goods Association And Recommendation Based On Hadoop

Tmall Supermarkets express boxes into the streets with the entity supermarkets,which was full of customers,virtually empty.This paper hopes to designs the corresponding data model for different data application through the analysis of a supermarket data,by using the Hadoop,which is an efficient and rapid data analysis platform.For example,put on the shelf products based on association rules or set a recommendation system based on the content recommendation algorithm,in order to put forward a kind of supermarket marketing decision scheme,which can improve entities in supermarket sales and earnings.First of all,this paper studied the correlation between the distributed platform Hadoop and mining supermarket data.Large foreign supermarkets had used Hadoop analysis example of supermarket sales data to predict customer behavior,and there were examples of large-scale data analysis in domestic market.At the same time,the supermarket had required hardware facilities to establish rapidly this platform.Then,taking supermarket sales receipts data and member customers shopping data as the research object,carried association rules analysis out on the items and evaluated value of association rules by profit from kinds of sales profit.Meanwhile,the recommendation system generated cost-effective recommendations as analysing the customer data,based on the value of goods and profit.In this paper,the data model was put forward by experiment,and the experimental results were compared with traditional data analysis tools.The conclusion confirmed that the theory was rational and more advantageous,which had certain extension value in actual supermarket marketing.