diplomsko delo
Aleksej Miloševič (Author), Vili Podgorelec (Mentor), Sašo Karakatič (Co-mentor)

Abstract

V pričujočem diplomskem delu sta analizirani in primerjani splošnonamenski platformi za podatkovno rudarjenje RapidMiner in Weka. V uvodnem delu diplomskega dela so razložene osnove strojnega učenja in podatkovnega rudarjenja ter podrobneje definirane metode dela, ki so uporabljene v praktičnem delu. Primerjava je razdeljena na teoretični in eksperimentalni del. V teoretičnem delu so na podlagi definirane metodologije identificirane pomembne lastnosti orodij in primerjane med seboj, v eksperimentalnem delu pa sta primerjani točnost in F-Mera implementacij algoritmov k-najbližjih sosedov, Naključni gozdovi in Naivni Bayes. S pomočjo statističnih testov je bilo ugotovljeno, da se nobena izvedenka algoritma od drugega statistično pomembno ne razlikuje.

Keywords

strojno učenje;klasifikacija;primerjava platform;RapidMiner;Weka;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher: A. Miloševič
UDC: 004.65(043.2)
COBISS: 19991062 Link will open in a new window
Views: 1467
Downloads: 213
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: ANALYSIS AND COMPARISON OF DATA MINING PLATFORMS RAPIDMINER AND WEKA
Secondary abstract: The following thesis analyses and compares two general-purpose platforms for data mining, RapidMiner and Weka. The introductory part of this diploma thesis describes the basics of machine learning and data mining as well as the specifically defined work methods, which are used in the experimental part. The comparison is divided into the theoretical and the empirical part. In the theoretical part the important characteristics of the tools are identified and compared on the basis of the defined methodology, whereas in the empirical part the accuracy and the F-measure of implementations of the algorithms K Nearest Neighbor, Random Forest and Naive Bayes are compared. Using appropriate statistical tests, it was found that no version of the algorithm significantly differs from another.
Secondary keywords: machine learning;classification;comparison;RapidMiner;Weka;
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, 77 str.
ID: 9166091