Building a Business Intelligence System for AUB (second edition)

Studenteropgave: Kandidatspeciale og HD afgangsprojekt

  • Trien Huy Ly
  • Jacob Mogensen
  • Kristian Reesen Skouboe
4. semester, Datalogi, Kandidat (Kandidatuddannelse)
This report describes the development of a business intelligence (BI) system for AUB. The foundation of the BI system is a data warehouse. It can be used for finding association rules, getting statistics, and finding book recommendations.
At first, AUB''s requirements are described. Based on the requirements the data warehouse for AUB is developed.
We have implemented our own algorithm (LIQ) for finding association rules. Furthermore, we have made an hybrid out of LIQ and Apriori to improve performance by benefiting the strengths of both worlds.
In addition, we have implemented a recommendation service based on both an item-based and a content-based approach in order to improve the quality of the recommendations.
The implementations for both association rule mining and the recommendation service are described along with performance tests of the implementations.
Finally, in order to demonstrate the statistical features, association rule mining, and recommendation service, we have implemented a data access web tool for this purpose.
Udgivelsesdatojun. 2004
ID: 61061395