Комбиниране на анализ на пазарната кошница с методи за клъстеризиране и RFM анализ
Combining market basket anajysis with clustering and RFM analysis

Author(s): Yanka Aleksandrova
Subject(s): Economy
Published by: Икономически университет - Варна
Keywords: market basket analysis; clustering; RFM analysis; association rules
Summary/Abstract: The association analysis as a data mining technique has been used for a long time for market basket analysis. The application of association analysis in its classic however form does not take into account important customer and transactions features. The paper proposes an approach for the application of association analysis along with clustering and RFM analysis in order to overcome some limitations of the market basket analysis. The results show that the combination of association analysis with other data mining algorithms and RFM analysis provides better opportunities to extract knowledge about jointly purchased products/services compared to applying only an association analysis. Two main groups of variants with two subsets are found for the application of the clustering and market basket analysis – association analysis on sales for customer clusters and clustering after selection by generated association rules. The proposed approaches are implemented in Microsoft SQL Server Analysis Services and Visual Studio environment and results are discussed.

• Details
• Contents