Analiza socialnega omrežja na podlagi zapisov podatkov o klicih

magistrsko delo

Rok Kozel (Author), Dušan Gleich (Mentor), Uroš Župerl (Mentor), Iztok Šerbec (Co-mentor)

Abstract

V tem delu se ukvarjamo z analizo podatkov, ki jih generira končni uporabnik v mobilnem omrežju. Pokazati želimo možnost priprave podatkov za modeliranje ter predlagati nov način nastavitve optimalnih parametrov modela. Iz podatkov smo izluščili informacije s pomočjo metod rudarjenja, kjer smo želeli pridobiti informacije o aktivnih in neaktivnih uporabnikih v določenem deležu populacije. Cilj je preizkusiti metode rudarjenja pri določeni topologiji omrežja in preveriti ustreznost modelov za določenega mobilnega operaterja. Predlagamo še pripravo izhodnih podatkov modela ter rezultate vizualiziramo, ovrednotimo in jih primerjamo s prehodnimi deli.

Keywords

mobilno omrežje;analiza omrežja;podatkovna znanost;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2016
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	R. Kozel
UDC:	004.6:004.7(043.2)
COBISS:	19828502
Views:	946
Downloads:	107
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Social Network Analysis on Basis of Call Detail Records
Secondary abstract:	In this dissertation we are dealing with analysis of the data generated by end user in mobile network. We want to show possibility of preparing the data for modeling and propose a new way of setting optimal parameters for the model. From data we extracted information using data mining methods, where we want to obtain information on active and inactive user in specified proportion of population. The aim is to test mining methods in particular network topology and verify if our models are appropriate for particular mobile operator. We suggest further preparation data from model output and visualize results, evaluate and compare them to previous work.
Secondary keywords:	mobile network;network analysis;churn;data science;
URN:	URN:SI:UM:
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Mehatronika
Pages:	XVIII, 73 str.
ID:	9151507