Web Usage Mining with SQL Server 2000

Data mining is the process of extracting valid and comprehensible information from conventional databases and using it to make crucial business decisions. It is one of the most exciting features of Analysis Services in SQL Server 2000. You can use data mining with either relational data sources or OLAP cubes.

In this article, author Denis Susac describes the design and implementation process for a data--mining framework that can be used as an addition to the IBuySpy portal or any other document management framework and dynamic website. It collects the history of users' interactions, enabling you to perform two tasks: clustering users according to the pages they visit frequently, and suggesting other pages of interest using a collaborative filtering approach. These techniques can also be applied to market basket analysis, recommendation engines for online stores, and other similar objectives. System requirements are SQL Server 2000 (Analysis Services), the .NET Framework, and IIS 5.0.