A Data Warehouse Cleansing Approach Based on Mathematical Association Rules

Abstract:

The alliance rules stated above based on the principle of data mining association rules provide a solution for detecting errors in the data sets. The errors are detected automatically. The manual intervention in the proposed algorithm is highly negligible resulting in high degree of automation and accuracy. The duplicity in the names field of the data warehouse has been remarkably cleansed and worked out. Domain independency has been achieved using the concept of integer domain which even adds on to the memory saving capability of the algorithm.

• Periodical: Advanced Materials Research (Volumes 490-495)
• Pages: 1878-1882
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.490-495.1878
• Online since: March 2012
• Authors: Yu Xiang Song
• Keywords: Association Rule, Data Cleansing, Data Warehouse

References

[1] J. Karlsteen, Automation of metadata updates in a time critical environment, (2006).

[2] P. Poonniah, Data Warehousing Fundamentals- A comprehensive guide for IT professionals, Ist ed., 81-265-0919-8, Glorious Printers: New Delhi , India, (2006).

[3] B. Palace, Data mining, http: /www. anderson. ucla. edu. [Online], 1996, http: /www. anderson. ucla. edu/faculty/jason. frand/teacher/technologies/palace/datamining. htm.

[4] C. Kelley, Best Uses of data warehoue, http: /www. itworld. com. [Online], 2003, http: /www. itworld. com/nl/db_mgr/, (2003).

[5] T. Redman, The Impact of Poor Data Quality on the Typical Enterprise, Communications of the ACM, Vol. 41. 8. 02, (1998).

[6] E. Rahm, H. H Do, Data Cleaning: Problems and Current Approaches, University of Leipzig, Germany.

[7] A. Marcus J.I. Maletic, "Automated Identification of Errors in Data Sets, TR-CS-00-02, University of Memphis, (2002).