Grupiranje država prema njihovoj mjeri sreće klasterskom analizom

master's thesis

• Title (english): Grouping countries according to their measure of happiness by cluster analysis
• Author: Lucija Karačić
• Mentor: Anamarija Jazbec (mentor)
• Committee member: Anamarija Jazbec (predsjednik povjerenstva)
• Committee member: Hrvoje Šikić (član povjerenstva)
• Committee member: Miljenko Marušić (član povjerenstva)
• Committee member: Ana Prlić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Mathematics); Zagreb
• Defense date and country: 2020-12-01, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Mathematics

Abstract

U ovom radu opisana je klasterska analiza i njezina primjena na bazu podataka koja sadržava informacije o 146 država. Klasterska analiza je metoda grupiranja objekata u klastere. Svaki je klaster grupa objekata najveće sličnosti. Ovisno o vrsti objekata i cilju istraživanja, odabiremo odgovarajuću mjeru sličnosti ili različitosti (udaljenosti). Nakon toga odabiremo algoritam klasteriranja; razlikujemo hijerarhijske i nehijerarhijske algoritme. Korištena baza dobivena je iz Svjetskog izvještaja o sreći (World Happiness Report) za 2017. godinu. Svakoj državi pridružene su vrijednosti varijabla Mjera sreće, Ekonomija, Obitelj, Zdravlje, Sloboda, Darežljivost i Povjerenje. S obzirom na to da su sve varijable numeričke, kao mjera udaljenosti odabrana je euklidska udaljenost. Najprije je provedeno hijerarhijsko (aglomerativno) klasteriranje pomoćuWardove metode i metode maksimuma. Zatim je provedeno nehijerarhijsko klasteriranje, tj. algoritam k-sredina za k = 5; 6; 7. Za obradu i analizu podataka korišten je programski sustav SAS.

Abstract (english)

This paper describes cluster analysis and its application to a database containing information on 146 countries. Cluster analysis is a method of grouping objects into clusters. Each cluster is a group of objects with the greatest similarity. Depending on the type of objects and the goal of the research, we choose the appropriate measure of similarity or difference (distance). After that, it is necessary to select the clustering algorithm; there are two types, hierarchical and non-hierarchical algorithms. Used database was obtained from the 2017 World Happiness Report. Value of variables Measure of Happiness, Economy, Family, Health, Freedom, Generosity and Trust, are assigned to each country. Since all variables are numerical, Euclidean distance was chosen as a measure of distance. Firstly, hierarchical (agglomerative) clustering was conducted using the Ward method and the maximum method. Next, non-hierarchical clustering was made with k-means algorithm for k = 5; 6; 7. The SAS software system was used for data processing and analysis.

• Keywords: klasterska analiza; primjena na bazu podataka; Mjera sreće; World Happiness Report za 2017. godinu
• Keywords (english): cluster analysis; application to a database; Measure of Happiness; 2017 World Happiness Report
• Language: croatian
• URN:NBN: urn:nbn:hr:217:790783
• Study programme: Title: Mathematical Statistics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra matematike (magistar/magistra matematike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2021-02-22 13:52:28