The Research On The Application Of Web Data Mining In E-commerce

The application of Internet data is growing at an alarming rate.When the Intelligent mobile phone,tablet computer and the cloud space promoting,encourage the problem of data expansion speed is even more serious. The economic globalization demands that entrepreneurs have courage to distinguish themselves, which contains attracting customer’s attention,good serving to customer,achieving mutual benefit and win-win with customers to. The performance of the platform is the electronic commerce website using the internet.But,In this platform,there are still having issues.That is" There are millions of data,but we don’t find the value of data".Data as reform and opening up,needs circulation which should be shaped to the development of era and technical requirements.Because of refusing to data means a rejection of wealth.The "4V" era of data is coming.Namely the data of Volume,Vaviety),Velocity),Value).The business advertising of the portal site,the payment and settlement of the online bank,the search engines,the social networks and other electronic commerce of various types in the form of data are changing people’s life.For the storage of data in quantity and the complexity of data in increase,data analysts break through the predicament, find a feasible approach,dig out the value of the data in order to help the data owner to find some regularity from large amounts of data to assist in decision making.This method is the technology of data mining.Electronic commerce is the engine of the economy in the future. Application of data mining of recommendation in the electronic commerce is the best way for enterprise to display image and product all-round of the world,to looking for partners and to expand the scale of selling.In this paper,the application of data mining technology in the field of electronic commerce recommendation system emerging has been studied.This paper mainly completed the following work:Firstly, this paper systematically discussed the current research situation of electronic commerce,data mining and recommendation system at home and abroad.Secondly,this paper described the application of Web data mining technology in the electronic commerce enterprise.Thirdly,this paper expounded how to use algorithms and techniques in recommender systems. Fourthly,this paper improved the recommended algorithm of Apriori, and designed an electronic commerce recommendation system based on Web data mining.Recommendation system is the key part of this paper.First of all,this paper made a feasibility analysis to recommendation system before designing.Then the recommendation system been divided into three big modules to design.The three big modules are data access module, system architecture-application module,and user interaction module.The next step of the paper is to segment of this three big modules.This paper detailed design the data access module which contains the data collecting module and the data processing module.the system architecture-application module which includes the architecture of the OLAP system module and the data mining system module based on B/S service,the user interaction module which comprises the online interactive recommendation module and the pattern of application module.This paper used the improved Apriori algorithm in architecture-application module,actualized the deduction of association rules,determined the associated page,and formed recommended set.This paper showed the operation interface of the user interaction module,and displayed the personalized service in the recommendation system.Although data mining technology in e-commerce recommendation system can bring great economic value and interest, but it’s also a double-edged sword.When businesses collecte a large amount of data,they also face new challenges in aspect of data’s processing,using, storage and security.How to effectively protect consumer’s privacy security.How to really improve enterprise’s value using data mining.In the mobile Internet era,how to let much more data appear in unstructured form.The data mining still has a long way to go.