CURE: učinkovit algoritam za grupiranje velikih skupova podataka

undergraduate thesis

• Title (english): CURE: An Effective Algorithm for Clustering Large Datasets
• Author: Lovre Torbarina
• Mentor: Siniša Srbljić (mentor)
• Committee member: Siniša Srbljić (predsjednik povjerenstva)
• Committee member: Zvonimir Pavlić (član povjerenstva)
• Committee member: Adrian Satja Kurdija (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom završnom radu obrađuje se tehnika grupiranja skupova podataka s naglaskom na CURE algoritam.
Započinjemo definiranjem pojma grupe i udaljenosti među grupama.
Objašnjena je osnovna podjela strategija za grupiranje skupova podataka.
Obrađeno je nekoliko algoritama koji pretpostavljaju Euklidski prostor i očekivan broj grupa. Prvo obrađujemo hijerarhijski algoritam, zatim obrađujemo
k-means algoritam te naposljetku CURE algoritam. Za svaki od algoritama objašnjena je njegova prostorna i vremenska složenost, te je napisan detaljan pseudokod s dodatnim obrazloženjima. Rad zaključujemo eksperimentima nad nekoliko malih i velikih skupova podataka pomoću kojih ukazujemo na prednost CURE algoritma u odnosu na druga dva obrađena u konkretnim situacijama.

Abstract (english)

In this bachelor thesis, we discuss data clustering technique with emphasis on CURE algorithm. We start by defining notation of the group and the distance between the groups. Then, we describe the basic classification of data clustering strategies. We process several algorithms that suggest Euclidean space and that the expected number of groups is known in advance. First, we analyze the hierarchical clustering algorithm, then we analyze k-means algorithm and at the last CURE algorithm. For every described algorithm, space and time complexity is given so as the detailed pseudocode with additional explanations. We conclude the bachelor thesis with experiments on a couple of small and big data sets with the aim to show the advantages of the CURE algorithm over the other two in concrete situations.

• Keywords: k-means; hijerarhijsko grupiranje; CURE; grupiranje skupova podataka
• Keywords (english): k-means; hierarchical clustering; CURE; dataset clustering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:789792
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-14 01:39:08