About a Technique to Examine the Relation between Products and Customers

Hiroshi Noguchi

《Asia Pacific Management Review》 11卷6期 (2006/12) Pp. 365-370

https://doi.org/10.6126/APMR.2006.11.6.04

並列摘要

There is Market Basket Analysis as a technique to determine which products sell together; moreover, there is Decision Tree (Classification Tree) technique to specify the common character of the customers who purchase a certain product. And, on the other hand, positioning methods are frequently applied to detect the relation between some products and customers. In this paper, we'll take up Correspondence Analysis (CA) as a typical one in positioning methods for discussion. CA is a technique that summarizes the structure of data and makes perceptual mapping. The benefit of CA lies in its unique abilities for representing rows and columns, for example, products and purchasers, in joint space. Therefore, we think that it is the most useful technique to examine the relation between products and purchasers. So, we'll show the improved analysis in order to examine the relation between products and the purchasers more delicately.

並列關鍵字

Market Basket Analysis ； Tree Analysis ； Correspondence Analysis ； Chi-square Distance Value ； Distant Measure

參考文獻

Anderson, T. W.(1963).A symptotic theory for principal component analysis.Annual Mathematics Statistics.34,122-148.

Benzecri, J. P.(1969).Statistical analysis as a tool to make patterns emerge from data.(Methodologies of Pattern Recognition).

Fisher, R. A.(1940).The precision of discriminate functions.Annal Eugen.10,422-429.

Gabriel, K. R.(1971).The biplot graphic display of matrixes with application to principal component analysis.Biometrika.8,433-467.

Greenarce, M. J.(1984).Theory and Application of Correspondence Analysis.London:Academic.