diplomsko delo
Iva Flisar (Author), Vili Podgorelec (Mentor), Sašo Karakatič (Co-mentor)

Abstract

Strojno učenje je pojem, tesno povezan s podatkovnim rudarjenjem, saj s pomočjo učnih algoritmov iščemo vzorce v podatkih. V diplomskem delu smo predstavili in opisali različne učne algoritme, ki se uporabljajo v procesu podatkovnega rudarjenja. Naš glavni cilj je bila predstavitev različnih metrik ovrednotenja učnih algoritmov. V ta namen smo v praktičnem delu diplomske naloge z različnimi metrikami ovrednotili učne algoritme. Eksperiment ovrednotenja smo izvedli na različnih podatkovnih množicah, ki smo jih razdelili z dvema različnima tipoma razdelitve – navzkrižno validacijo ter z metodo razdelitve.

Keywords

strojno učenje;klasifikacija;učni algoritmi;ovrednotenje algoritmov;metrike ocenjevanja;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher: I. Flisar
UDC: 004.8.021:004.6(043.2)
COBISS: 20177942 Link will open in a new window
Views: 1505
Downloads: 235
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Methods for the evaluation of machine learning algorithms
Secondary abstract: Machine learning is a concept closely related to data mining; we are looking for patterns in data, by using learning algorithms. In our thesis, we presented and described various learning algorithms that are used for data mining. Our main objective was to present different evaluation metrics of learning algorithms. We evaluated the learning algorithms in the practical part of the thesis. The experiment of evaluation was performed on different datasets which were divided with two different dividing methods - the cross validation and the holdout method. 
Secondary keywords: machine learning;classification;learning algorithms;evaluation algorithms;evaluation metrics;
URN: URN:SI:UM:
Type (COBISS): Bachelor thesis/paper
Thesis comment: Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages: X, 51 f.
ID: 9161492