Research Of Web Click Stream Based On Data Warehouse

Huge data have been produced everyday by Web. With the increase of the network information in many field common store of the Database could not satisfy the client's demand. In this condition, data warehouse gets more development. Data warehouse is gathering of data which is facing-topic, integration, changing continuously by time and steady. It is used to hold out process of decision-making in management. The different between Date Base and Data Warehouse is that Date Base system faces transaction processing, but Data Warehouse system faces analysis.Now the integration of Click stream data and information of clients has become foreland of date-analyze of web. Click stream data warehouse is used to analyze Click stream data. Click stream data warehouse is a facet of development of the technology of data warehouse. It includes dimensional molding method of date warehouse, ETL design of click stream data warehouse, implementation of Data warehouse and OLAP technology.At the same time, the application is increasingly divided into two types: On-Line Transaction Processing (OLTP) and On-Line Analytical Processing (OLAP). OLTP is mainly used for the application service of an organization. OLAP is used for the decision-supporting and analysis of the high level management. It is the future of information technology.In this article, first, author picks-up Click stream data and handle them. Then author makes sure of dimensionality and fair of data. Next, data is been extracted, transformed, cleaned and load. Last author designs and analyses the click stream data warehouse, and uses the case of education-net's click stream molding.First, collection and pretreatment in click stream data, in order to collect click information, collecting date in layer of server. We used the combination of Maximal ForwardReferenceModel and Time WindowModel in recognizing of User Transaction.Second, basic goal and dimensionality design of Education data warehouse, it included date market, dimension and measurement. Finally, ETL design of data warehouse and display multi-dimensional dataset and devise of OLAP technology after implemented of the data warehouse. We used DTS service in ETL and DMX method in OLAP, and also discuss some efficient technology in finding.