Application of data-mining techniques in customer segmentation

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F60460709%3A41110%2F11%3A51189" target="_blank" >RIV/60460709:41110/11:51189 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Application of data-mining techniques in customer segmentation
Original language description
The paper discusses the possibilities of using the various methods of customer segmentation by mining the information in the databases. The data-mining model was constructed from approximately 60 thousand transaction records. Only food records were selected for the analysis. The partial objective of the paper was to examine the possibilities of data preparation such as restructuralization, logarithmic transformation, and dimensionality reduction. The main goal of the paper was to find meaningful patterns in the analyzed data and identify clusters of customers with similar behavior and needs. The segmentation was realized by various data mining techniques as follows: K-means clustering, Two Step clustering, and unsupervized algorithm based on neural networks called Self-Organizing Maps. The quality of results was evaluated by the Silhouette measure, which combines the principles of clusters separation and cohesion. After that the detailed analysis of the final segments was done.
Czech name
—
Czech description
—

Publication year
2011
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
AGRARIAN PERSPECTIVES Proceedings of the 20th International Scientific Conference
ISBN
978-80-213-2196-0
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
403-408
Publisher name
PEF ČZU v Praze
Place of publication
Prague
Event location
Praha
Event date
Sep 13, 2011
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000338036100043

Similar results(10)