Tehnike klasteriranja pomoću teorije grafova

Ivančić, Ante (2014) Tehnike klasteriranja pomoću teorije grafova. = Clustering techniques using graph theory. Master's thesis (Bologna) , Sveučilište u Zagrebu, Fakultet strojarstva i brodogradnje, UNSPECIFIED. Mentor: Đukić, Goran.

[img] Text
Ivancic_Ante_Diplomski_Rad.pdf - Published Version
Restricted to Repository staff only Jezik dokumenta:Croatian

Download (4MB)

Abstract (Croatian)

U diplomskom radu opisane su tehnike klasteriranja pomoću teorije grafova te je provedena analiza na primjeru problema klasteriranja sličnih proizvoda, jednog od koraka tzv. korelacijskog odlaganja materijala u skladištu. Teorija grafova, kao grana diskretne matematike koja se bavi analizom grafova, tj. matematičkih struktura kojima se modeliraju mežđusobne veze izmeđžu objekata, predstavlja moćan alat pri rješavanju problema analize podataka koji su u svojoj suštini problemi kombinatorne optimizacije. Klasteriranje (nenadzirana klasifikacija) grana je strojnog učenja, čije tehnike služe prepoznavanju smislene strukture unutar skupa podataka putem grupiranja sličnih.

Abstract

The main topic of the thesis are graph-based clustering methods, which were described in great detail. A few of those methods were tested on the problem of clustering similar products, which presents a crucial step in the so-called ”correlated storage assignment strategy”. Graph theory, as a branch of discrete mathematics, which deals with graphs as mathematical structures used to model pairwise relations between objects, serves as a powerful tool for dealing with data analysis problems which are of combinatorial nature. Clustering (unsupervised classification) is a branch of machine learning that focuses on discovering hidden structure of the data by grouping a collection of objects into subsets or ”clusters”, such that those within each cluster are more closely related to one another than objects assigned to different clusters.

Item Type: Thesis (Master's thesis (Bologna))
Uncontrolled Keywords: teorija grafova; inženjerska logistika; nenadzirano učenje; klasteriranje; redukcija dimenzionalnosti
Keywords (Croatian): graph theory, engineering logistics, unsupervised learning, clustering, dimensionality reduction
Subjects: TECHNICAL SCIENCE > Mechanical Engineering
Divisions: 700 Department of Industrial Engineering > 710 Chair of Production Design
Date Deposited: 27 Nov 2014 11:35
Last Modified: 21 Apr 2020 10:47
URI: http://repozitorij.fsb.hr/id/eprint/3010

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year