Application of Data Mining Techniques for Proper Design of Knowledge Repository

This paper reports a basic assessment of the usage of a knowledge repository called "XYZ Knowledge Repository", maintained by a company PQR for its client XYZ. Subsequently, a data mining method for finding sequential patterns in the web usage of the users was used. This captures the level of association between the web pages through two parameters, support and confidence. Depending upon the values of these two parameters, concepts of strong association, weak association and no association are realised. In case of strong association we recommended the pages be grouped together; while in the case of weak association, a link may be provided.