Data Mining Techniques as a Support to Business Decision

Teohareva Filipova, Biljana (2012) Data Mining Techniques as a Support to Business Decision. Masters thesis, University Goce Delcev- Stip.

[img] Text
БИЛЈАНА ТЕОХАРЕВА-ФИЛИПОВА INFORMATIKA.pdf

Download (3728Kb)

Abstract

Data mining is a process of finding hidden regularities and connections among data. Base data mining can result in an increased accuracy of the current analyses. Data received in the process of data mining increase the efficiency and development of the business sector and help in bringing right, pro-active decisions which will influence the further development of a particular company. This Master's work implements data mining techniques in order to be used as a support to particular business decisions. Methods of decision trees, neuron networks, regression, Bayesian networks, associative rules and method of a spending basket were applied over data received from questionnaire leaves and fiscal accounts of customers in a Macedonian firm. These data were elaborated with the program tools SPSS and Clementine. Applying the data mining techniques brings answers to questions such as: what category of customers should be given loyalty cards, which factors influence on buying lower price products, which products are being bought together, which products are the most often being bought, which products would the customers like to be on an action etc. After issuing the loyalty cards, analysis of the sale has been done which shows that the firm profit is increased and it has been found what percentage of this profit belongs to loyal customers. A projection of the selling in the next period has been made on the basis of the sale in the previous period

Item Type: Thesis (Masters)
Subjects: Natural sciences > Computer and information sciences
Divisions: Faculty of Computer Science
Depositing User: Snezana Georgieva Siceva
Date Deposited: 07 Nov 2012 18:47
Last Modified: 14 Oct 2014 12:10
URI: http://eprints.ugd.edu.mk/id/eprint/616

Actions (login required)

View Item View Item