prikaz prve stranice dokumenta Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa
  Download
PDF 1.1 MB
undergraduate thesis
Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa
Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. urn:nbn:hr:200:048928
Džoić, Kristian
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Software Engineering
Chair of Programming Languages and Systems

Institutional repository: Faculty of Electrical Engineering, Computer Science and Information Technology Osijek

Cite this document

Džoić, K. (2024). Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa (Undergraduate thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:048928

Džoić, Kristian. "Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:048928

Džoić, Kristian. "Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:048928

Džoić, K. (2024). 'Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 28 December 2024, https://urn.nsk.hr/urn:nbn:hr:200:048928

Džoić K. Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2024 [cited 2024 December 28] Available at: https://urn.nsk.hr/urn:nbn:hr:200:048928

K. Džoić, "Utjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:200:048928

Metadata
TitleUtjecaj parametara algoritma SMOTE za rukovanje problemom neuravnoteženosti klasa
Title (english)Impact of the parameters of the SMOTE algorithm for handling class imbalance
AuthorKristian Džoić
MentorDražen Bajer (mentor)
Committee memberEmmanuel Karlo Nyarko (predsjednik povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Software Engineering)
(Chair of Programming Languages and Systems)
Osijek
Defense date and country2024-09-26, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
sažetak (hrvatski): U radu je istražen problem skupova podataka sa svojstvom neuravnoteženosti klasa koje otežava klasifikaciju manjinske klase, te mogućnosti rješavanja tog problema primjenom algoritma SMOTE, koji pruža preuzorkovanje podataka stvaranjem sintetičkih uzoraka manjinske klase s ciljem postizanja ravnoteže između klasa. Posebna pažnja posvećena je analizi utjecaja različitih postavki parametara algoritma SMOTE, uključujući broj susjeda i broj generiranih sintetičkih uzoraka, na performanse klasifikacijskih modela. Eksperimentalna analiza provedena je koristeći više klasifikatora, pri čemu su rezultati ocjenjivani F-1 i G-Mean mjerama. Pokazalo se da pažljivo odabrane postavke parametara mogu značajno poboljšati učinkovitost klasifikacije, dok nepravilne postavke mogu dovesti do prekomjerne prilagodbe podacima za treniranje i smanjenja performansi. Zaključeno je da se parametri algoritma SMOTE trebaju prilagoditi specifičnostima skupa podataka i klasifikatora kako bi se postiglo poboljšanje učinkovitosti klasifikacije.
Abstract (english)
abstract: The paper explores the issue of datasets with class imbalance, which complicates the classification of the minority class, and the possibilities of addressing this problem using the SMOTE algorithm , which enables data resampling by creating synthetic samples of the minority class with the aim of achieving balance between classes.. Special attention is given to analyzing the impact of different settings of SMOTE algorithm parameters, including the number of neighbors and the number of generated synthetic samples, on the performance of classification models. The experimental analysis was conducted using multiple classifiers, with results evaluated by the F-1 and G-Mean metrics. It was shown that carefully selected parameter settings can significantly improve classification performance, while incorrect settings can lead to overfitting to the training data and reduce performance. It is concluded that the parameters of the SMOTE algorithm should be adjusted to the specific characteristics of the dataset and the classification model to achieve classification efficiency improvements. s
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:048928
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-10-14 08:43:07