Design And Realization Of Data Mining System Base On Web

As to enterprise, how to survive and develop in the fierce commerce day by dayof the competition, strive for more customers seem extremely important. If we canknow what the customer needs, what is interested in, make key decision and put intopractice on this basis, could attract users to buy one’s own goods, obtain profits forenterprises. The data mining, as a kind of new developing commercial technology,become the key of solving this problem.Users such as browsing through webpage,etc., will leave a large number ofrecords by daily behavior. This seemingly useless information in the system of datamining is processed with analyzing, changing, extracting etc., can receive theimportant data helped to the business decision greatly.This text has studied the following contents:First of all, I have probed into the concept of the data mining, have introducedcourse and function of the data mining, and has carried on research to the datawarehouse. This data mining adopts web excavation especially, and has discussedrelevant rudimentary knowledge of the characteristic that web excavation and user’sbehavioral analysis in detail.Secondly, each function module of data mining is designed, and carry on researchto the system of the data mining. Some commonly used algorithms of detailedintroduction excavate the system to classify to the data. Make an introduction toalgorithms such as the association rules, the clustering algorithm and the regressionanalysis algorithm,etc., especially, we structure the web algorithm of this text used.Finally, This text design the system of data mining, build up modules of thesystem architecture, include the data reading module, preconditioning module andinterest mining module, and then debug systematic function, check with precision andrecall that leads these2indexes to imitate the assessment completely, provesystematic dependability, and demonstrate some data excavated out systematically.Originally design through drawing a large number of records which users left inthe webpage in XML file, as the basic data source, has set up the general model thatthe data mining, through algorithms such as the association rules, the clusteringalgorithm and the regression analysis algorithm,etc., work out the universal law thatusers surfed the net in the behavior, analyze the demand in user’s subconscious, offer effective support for making a decision to help commercialized operation inenterprises.