Clustering methods are recently so advanced elaborated algorithms for large collection data analysis that they have been already included today to data mining methods. Clustering methods are nowadays larger and larger group of methods, very quickly evolving and having more and more various applications. In the article, our research concerning usefulness of clustering methods in customer segmentation to manage advertisement campaign is presented. We introduce results obtained by using four selected methods which have been chosen because their peculiarities suggested their applicability to our purposes. One of the analyzed method - k-means clustering with random selected initial cluster seeds gave very good results in customer segmentation to manage advertisement campaign and these results were presented in details in the article. In contrast one of the methods (hierarchical average linkage) was found useless in customer segmentation. Further investigations concerning benefits of clustering methods in customer segmentation to manage advertisement campaign is worth continuing, particularly that finding solutions in this field can give measurable profits for marketing activity.
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.