Show simple item record

dc.contributor.advisorLepojević, Vinko
dc.contributor.otherĐorđević, Vera
dc.contributor.otherKalinić, Zoran
dc.creatorMilanović, Marina B.
dc.date.accessioned2019-05-30T12:56:36Z
dc.date.available2019-05-30T12:56:36Z
dc.date.issued2019-01-15
dc.identifier.urihttp://eteze.ni.ac.rs/application/showtheses?thesesId=6694
dc.identifier.urihttps://fedorani.ni.ac.rs/fedora/get/o:1573/bdef:Content/download
dc.identifier.urihttp://vbs.rs/scripts/cobiss?command=DISPLAY&base=70052&RID=513884508
dc.identifier.urihttp://nardus.mpn.gov.rs/123456789/11145
dc.descriptionThe development of information technology and, consequently, rapid increase of the available amount of data have contributed to the fact that data mining, as a key component of a wider interactive, iterative and creative process of knowledge discovery from data, is of great importance in economic research. The basic idea of data mining is reflected in the efficient and effective identification of regularities, that are hidden in large sets of (multidimensional) data stored in information repositories, using software-supported methods and algorithms. Taking into account the foregoing, in this doctoral dissertation, the most important theoretical-methodological aspects of data mining approaches in data analysis are examined, as well as its applicative possibilities in the field of studying economic phenomena. In this sense, trends in the modern economy, from the perspective of the growing role of data (as a usable resource for generating values), fundamental terminology related to the concept of data mining as well as the positive and negative contexts of its application, have been analyzed. The tasks of discovering knowledge from data, in function of creating a data mining model, are viewed through the prism of a wide range of methodological procedures for their implementation. Special attention has been dedicated to the relationship between data mining and statistics, as a science that traditionally deals with the discovery of regularities from data. The results of the research indicate the great potential of integrated implementation of statistical and data mining approaches in finding innovative methodological solutions to specific problems and, at the same time, suggest the need for mutual adaptation and modification of basic paradigms of data analysis on both sides. Empirical part of the dissertation points out innovative conceptually-methodological frameworks for analyzing data from time series of stock exchange indices and survey data on service users. Empirical results, as the exact knowledge extracted from data, confirm the importance of implementing data mining analysis in the problem contexts of economics, business economics and management. The conducted research represents a suitable basis for profiling future research orientations in the field of data mining application concerning the study of economic phenomena.en
dc.formatapplication/pdf
dc.languagesr
dc.publisherУниверзитет у Нишу, Економски факултетsr
dc.rightsAutorstvo-Nekomercijalno-Bez prerade 3.0 Srbija (CC BY-NC-ND 3.0)
dc.sourceУниверзитет у Нишуsr
dc.subjectEkonomski podaci, znanje, zakonitosti, data mining, statistikasr
dc.subjecteconomic data, knowledge, regularities, data mining, statisticsen
dc.titleIzvođenje zakonitosti iz ekonomskih podataka primenom data mining pristupasr
dc.typePhD thesis


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record