diplomsko delo
Aljaž Heričko (Author), Vili Podgorelec (Mentor)

Abstract

V diplomski nalogi predstavimo osnovne značilnosti klasifikacijskih pristopov, ki temeljijo na podatkovni gravitaciji ter predlagamo nov klasifikacijski algoritem in klasifikator, ki temelji na principu podatkovne gravitacije. Učinkovitost predlaganega klasifikatorja v smislu točnosti razvrščanja smo ovrednotili na naboru sedmih standardnih podatkovnih zbirk in rezultate primerjali z drugimi uveljavljenimi klasifikacijskimi algoritmi. Na osnovi rezultatov eksperimentalne študije lahko sklepamo, da predlagani klasifikator, podobno kot nekateri drugi na podatkovni gravitaciji temelječi pristopi, zagotavljajo zadovoljive in primerljive rezultate nad standardnimi testnimi podatkovnimi množicami.

Keywords

umetna inteligenca;strojno učenje;podatkovna gravitacija;tehnike klasifikacije;orodje Weka;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher: [A. Heričko]
UDC: 004.6:004.896(043.2)
COBISS: 18540822 Link will open in a new window
Views: 1535
Downloads: 137
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: DATA GRAVITATION BASED CLASSIFICATION
Secondary abstract: In this diploma work a new classification algorithm and a corresponding classifier named Simple Data Gravitation based Classifier (SDGC) is proposed. The efficiency of the proposed classifier was evaluated using seven standard data sets and compared to the results of selected well-known classification algorithms. Results of the conducted experimental study illustrate that the proposed classifier gives promising results on standard test data.
Secondary keywords: artificial intelligence;machine learning;data gravitation;classification methods;Weka;
URN: URN:SI:UM:
Type (COBISS): Bachelor thesis/paper
Thesis comment: Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko
Pages: IX, 35 f.
ID: 8731422