Research And Implementation Of Intelligent Chain Convenience Store System Based On Improved Apriori Algorithm

With the economic development coming into a stable period,chain retail industry also abandon the high growth and get into a "personalized" era.It is one of the effective means to promote the chain retail industry skip out of the bottleneck situation that we put out the personalized requirements,promote the changing of retail business strategy,and build a suitable chain of retail information systems that meet the individual requirements of customers in different regions.Therefore,this paper studied and accomplished the chain convenience store intelligent management information system based on cloud database(RDS)for the retail business characteristics of chain convenience stores,the association rules of commodity sales and the individual requirements of customers.First of all,this paper presents a data storage model based on cloud database.The exchange and storage of the chain convenience store information system data are in the headquarters of the server.However,the headquarters business data gradually increased when time flows,so the headquarters server performance bottlenecks and expansion bottlenecks with data security have become a problem for system to solve.We set up the headquarters database in the cloud database,using cloud database to storage,backup,expand and design security the convenience store's data.In this way,we not only save the maintenance costs of the server,but also enhance the security of data security,and also the requirements of scalability.Second,this paper designs and implements a real-time data transfer between stores and headquarters.There may emerge network congestion and data inconsistencies during the process of data transmission between each convenience store and the headquarters.We used the postgre_fdw expand and finish cross-database operations between each PostgreSQL database or cloud database database,Thus we can effectively avoiding network congestion or database connection anomalies which caused by the increasing data.Third,this paper improves the Apriori algorithm and proposes a model of intelligent chain convenience store system based on data association rule mining.During the process of algorithm Apriori mining the data,there are some issues such as there will produce a large number of candidate sets,and in the calculation of candidate set support degree we need to scan the entire database which resulting in the low performance of the algorithm.As for these problems,this paper proposes a strategy to optimize the Apriori algorithm from the three aspects,database scan optimization,frequent item sets pre-connection optimization and frequent item sets connection optimization.This strategy uses the database support matrix to store frequent 1-item sets,frequent 2-item sets of support,and uses the binary support degree vector transaction set to reduce the number of transactions and comparison to avoid too much accession to the original database.Therefore we can effectively improve the performance of the Apriori algorithm.In addition,the improvement strategy can extract and deal with the previous sales data of the chain of convenience stores,and then use the optimized Apriori algorithm to digging the buying habits of the convenience store customers'.And then it will get the association rules among the products and apply the rules to the system procurement,distribution,promotion and other modules,and design a set of the chain convenience store management system intelligent module which is based on commodity relationship.Finally,after the experiments tests and the preliminary operation in the enterprise,we verified the effectiveness and advance of the chain convenience store intelligent information system we researched and developed.