izpis_h1_title_alt

Analiza soizbirnosti predmetov na Fakulteti za računalništvo in informatiko
ID SMOLE, PETRA (Author), ID Šubelj, Lovro (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (5,63 MB)
MD5: 3C2FC7613BB220329FCB8825064CAB2E
PID: 20.500.12556/rul/f9f8c144-ca15-417d-a221-11c80c30f33a

Abstract
Analiza omrežij in s tem številne omrežne mere omogočajo vpogled v lastnosti omrežja ter imajo pri razumevanju le-tega pomembno vlogo. V diplomskem delu je podana analiza omrežja visokošolskega in magistrskega študijskega programa Fakultete za računalništvo in informatiko. Namen naloge je poiskati skupine predmetov glede na to, kako jih študentje soizbirajo. To smo dosegli z uporabo petih že uveljavljenih pristopov, namenjenih iskanju skupin v omrežju in s pomočjo prosto dostopnih programskih orodij Gephi, Visone ter Pajek. Za iskanje skupin uporabimo pristop jedrno-obrobne zgradbe, optimizacijo modularnosti, izmenjavo oznak, bločno modeliranje ter stohastično bločno modeliranje. Rezultate, dobljene s pravkar predstavljenimi pristopi, primerjamo in v zadnjem delu podamo interpretacijo tistih, ki so z našega stališča zanimivi ter uporabni. Predmeti posameznega študijskega modula oziroma predmeti izvajalcev istega laboratorija se ne dodelijo le eni od skupin, vendar pa se v nekaterih primerih temu dobro približajo.

Language:Slovenian
Keywords:analiza omrežij, skupine vozlišč, skupnosti vozlišč, jedrno-obrobna zgradba
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2016
PID:20.500.12556/RUL-85187 This link opens in a new window
Publication date in RUL:14.09.2016
Views:1325
Downloads:551
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Analysis of course co-enrollment at the Faculty of Computer and Information Science
Abstract:
Network analysis and consequently numerous network measurements allow an insight into the characteristics of network and therefore have an important role in its understanding. In this thesis the analysis of the professional and masters study programme of the Faculty of Computer and Information Science is conducted. The purpose is to find groups of courses based on how the students co-enroll them. We have achieved this by using five already established approaches intended for revealing characteristic groups in a network and with the help of widely-accessible software tools such as Gephi, Visone and Pajek. For revealing groups we use the approaches for core-periphery structure, modularity optimization, label propagation, blockmodeling and stochastic blockmodeling. The results obtained with the introduced approaches are compared and, in the last part of the thesis, we provide the interpretation of those we find interesting and useful from our point of view. Courses of a particular study module or courses of instructors from the same laboratory are not assigned to just one of the groups, but they do, however, get close in some cases.

Keywords:network analysis, groups of nodes, communities, core-periphery structure

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back