| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:Gradnja uravnoteženih evolucijskih klasifikacijskih dreves : magistrsko delo
Authors:ID Lahovnik, Tadej (Author)
ID Karakatič, Sašo (Mentor) More about this mentor... New window
ID Podgorelec, Vili (Comentor)
Files:.pdf MAG_Lahovnik_Tadej_2025.pdf (2,85 MB)
MD5: FCCBC9CD52E6C12B19DE37D1FD0993FA
 
Language:Slovenian
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:Uspešnost odločitvenih dreves temelji na predpostavki, da učni podatki za vsak razred vključujejo enako količino informacij. Pri nesorazmerni porazdelitvi razredov so klasifikatorji pristransko usmerjeni k večinskim razredom. Zaradi majhnega števila vzorcev manjšinskih razredov klasifikatorji niso zmožni ustreznega usvajanja znanja, kar vodi do slabšega posploševanja in prekomernega prileganja. V okviru zaključnega dela smo razvili več algoritmov za gradnjo uravnoteženih evolucijskih dreves, ki se osredotočajo na reševanje izzivov, povezanih z nesorazmerno porazdelitvijo razredov. Rezultati eksperimenta kažejo, da uravnoteženost evolucijskih dreves ne prispeva k izboljšanju klasifikacije v primerjavi s tradicionalnimi metodami.
Keywords:evolucijski algoritem, odločitvena drevesa, klasifikacija, neuravnoteženi podatki
Place of publishing:Maribor
Place of performance:Maribor
Publisher:[T. Lahovnik]
Year of publishing:2024
Number of pages:1 spletni vir (1 datoteka PDF (XII,74 str.))
PID:20.500.12556/DKUM-91397 New window
UDC:004.021:575.82(043.2)
COBISS.SI-ID:227265539 New window
Publication date in DKUM:06.02.2025
Views:147
Downloads:102
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share



Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Link:http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:18.12.2024

Secondary language

Language:English
Title:Construction of balanced evolutionary classification trees
Abstract:The performance of decision trees relies on the assumption that the training data for each class includes the same amount of information. When the distribution of classes is imbalanced, the classifiers are biased towards the majority classes. Due to the small number of minority class samples, the classifiers cannot adequately extract knowledge, leading to poor generalisation and overfitting. As part of our final work, we developed several algorithms for building balanced evolutionary trees that address the challenges associated with imbalanced class distributions. Experimental results show that balanced evolutionary trees do not contribute to classification improvement compared to traditional methods.
Keywords:evolutionary algorithm, decision trees, classification, imbalanced data


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica