dc.contributor.advisor | Stanković, Milena | |
dc.contributor.other | Stoimenov, Leonid | |
dc.contributor.other | Stojković, Suzana | |
dc.contributor.other | Stanković, Miomir | |
dc.contributor.other | Milovanović, Slavoljub | |
dc.creator | Marković, Ivana P. | |
dc.date.accessioned | 2018-07-05T09:38:18Z | |
dc.date.available | 2018-07-05T09:38:18Z | |
dc.date.available | 2020-07-03T16:02:53Z | |
dc.date.issued | 2018-05-03 | |
dc.identifier.uri | http://eteze.ni.ac.rs/application/showtheses?thesesId=5861 | |
dc.identifier.uri | https://nardus.mpn.gov.rs/handle/123456789/9591 | |
dc.identifier.uri | https://fedorani.ni.ac.rs/fedora/get/o:1491/bdef:Content/download | |
dc.identifier.uri | http://vbs.rs/scripts/cobiss?command=DISPLAY&base=70052&RID=533967510 | |
dc.description.abstract | The aim of the research presented within this doctoral dissertation is
to develop a feature selection methodology through integrating
domain-specific knowledge by applying mathematical methods of
decision-making, to improve the feature selection process and the
precision of supervised machine learning methods for predictive
modeling of time series.
To integrate domain-specific knowledge, a multi-criteria decision
making method is used, i.e. an analytical hierarchical process proven
to be successful in numerous studies carried out to date. This
approach was selected because it allows the selection of a set of
factors based on their relevance, even in the case of mutually opposite
criteria.
In predicting the movement of time series, the possibility of
integrating feature relevance into support vector machines to improve
their prediction accuracy was studied.
The proposed methodology was applied as a feature-selection method
for the predictive modelling of movement of financial time series.
Unlike existing approaches, where the feature selection method is
based on a quantitative analysis of the input values, the proposed
methodology carries out a qualitative evaluation of the attributes in
relation to the prediction domain and represents a means of
integrating a priori knowledge of the prediction domain. | en |
dc.format | application/pdf | |
dc.language | sr | |
dc.publisher | Универзитет у Нишу, Електронски факултет | sr |
dc.rights | openAccess | en |
dc.rights.uri | https://creativecommons.org/licenses/by-nc-nd/4.0/ | |
dc.source | Универзитет у Нишу | sr |
dc.subject | izbor atributa | sr |
dc.subject | Feature selection | en |
dc.subject | Weighted kernel function | en |
dc.subject | Predictive modeling | en |
dc.subject | Time series | en |
dc.subject | težinska kernel funkcija | sr |
dc.subject | prediktivno modelovanje | sr |
dc.subject | vremenske serije | sr |
dc.title | Izbor atributa integracijom znanja o domenu primenom metoda odlučivanja kod prediktivnog modelovanja vremenskih serija nadgledanim mašinskim učenjem | sr |
dc.type | doctoralThesis | en |
dc.rights.license | BY-NC-ND | |
dcterms.abstract | Станковић, Милена; Стоименов, Леонид; Стојковић, Сузана; Станковић, Миомир; Миловановић, Славољуб; Марковић, Ивана П.; Избор атрибута интеграцијом знања о домену применом метода одлучивања код предиктивног моделовања временских серија надгледаним машинским учењем; Избор атрибута интеграцијом знања о домену применом метода одлучивања код предиктивног моделовања временских серија надгледаним машинским учењем; | |
dc.identifier.fulltext | https://nardus.mpn.gov.rs/bitstream/id/52370/Markovic_Ivana_P.pdf | |
dc.identifier.fulltext | http://nardus.mpn.gov.rs/bitstream/id/52369/Disertacija.pdf | |
dc.identifier.fulltext | http://nardus.mpn.gov.rs/bitstream/id/52370/Markovic_Ivana_P.pdf | |
dc.identifier.fulltext | https://nardus.mpn.gov.rs/bitstream/id/52369/Disertacija.pdf | |
dc.identifier.rcub | https://hdl.handle.net/21.15107/rcub_nardus_9591 | |