K- Frequently Based On Association Rules And Clustering Algorithms In Embedded Systems

Since the1960s, with the rapid development of computer, networking and communications,the popularity of the Internet and applications e-commerce, office automation, management information systems, making people accumulate more and more data. But if you want to get useful information and knowledge, the traditional means of data analysis has been difficult to meet the demand, which resulted in a contradiction, that the rapid data generation, storage, search, and relatively backward data analysis methods contradiction between.while in the embedded field, with the steady development of science and technology, including embedded microcomputer hardware performance have been developed by leaps and bounds.Therefore, in order to make full use of existing hardware development opportunities and improve the efficiency of data mining, this paper based on the retail industry as the research background developing a small data mining system on the embedded micro-computer, the system uses the association rules mining in data mining technology as a means of analysis, due to the classic Apriori algorithm is inefficient and difficult to promote in embedded devices, so this paper puts forward a new algorithm of association rules:Based on the K-frequent association rule clustering algorithm. The new algorithm effectively avoids the Apriori algorithm efficiency is not high faults. Application of the algorithm design of embedded micro systems that can perform efficient data mining processing, mining the association rules, formulates the commodity shelf design, goods storage arrangement and analysis to provide important information for retailers buying patterns. This paper implements effective combination of embedded technology and data mining technology, makes the data mining possible to run in small embedded devices.