No public access
master's thesis
7   0
Customer churn prediction for telecommunication service provider by analyzing implicit social network based on customer activity
Ana Marija Galić (2015)
University of Zagreb
Faculty of Electrical Engineering and Computing
Metadata
TitlePredviđanje gubitka korisnika iz perspektive davatelja telekomunikacijskih usluga analizom implicitne društvene mreže definirane korisničkim aktivnostima
AuthorAna Marija Galić
Mentor(s)Marko Banek (thesis advisor)
Abstract
Rad se bavi uvodom u analizu društvenih mreža u svrhu predikcije churna (odlaska korisnika) u telekomunikacijskim mrežama. Dan je uvid u analizu društvenih mreža (SNA): što je to, kako se i zašto koristi, kako je nastala i koja je razlika između SNA i dubinske analize podataka. Napravljen je pregled najosnovnijih mjera korištenih u predikciji churna, objašnjeno je što su to CDR podaci e je dan primjer na pravom skupu podataka. Na temelju tog skupa promatra se optimalna mjera za predikciju churna s obzirom na utjecaj ostalih čvorova u mreži na promatrani čvor. Također, objašnjeno je što su čvorovi, gdje ima mjesta za napredak i razvoj te je dan primjer alata kojim se analiza može raditi.
Keywordssocial network analysis churn marketing leaders and followers datamining telco network eigenvector indegree and outdegree CDR dataset weighted and non-weighted graphs
Parallel title (English)Customer churn prediction for telecommunication service provider by analyzing implicit social network based on customer activity
Committee MembersMarko Banek (committee chairperson)
Marko Banek (committee chairperson)
Krešimir Pripužić (committee member)
Zoran Skočir (committee member)
Luka Humski (committee member)
Boris Vrdoljak (committee member)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
PlaceZagreb
StateCroatia
Scientific field, discipline, subdisciplineTECHNICAL SCIENCES
Computing
Study programme typeuniversity
Study levelgraduate
Study programmeComputing
Academic title abbreviationmag.ing.comp.
Genremaster's thesis
Language Croatian
Defense date2015-07-03
Parallel abstract (English)
This thesis gives an introduction to social networks analysis in predicting Churn (users leaving supplier) in telecommunication networks. An insight into the social network analysis (SNA) is given: what it is, how and why is it used,and what is the difference between SNA and data mining. A review of the most basic measures used in the churn prediction is given, as well as an example of SNA on the real life dataset. Based on that set, the optimal measures for churn prediction is given. Conclusion on where there is room for improvement and development is given, as well as an example of a good tool that can be used
Parallel keywords (Croatian)mjere telekomunikacijska mreža analiza društvene mreže teorija grafova grafovi R RStudio svojstveni vektor (eigenvector) gustoća mreže težinski grafovi bestežinski grafovi odlazak korisnika iz mreže CRISP-DM dubinska analiza podataka CDR podaci o pozivima
Resource typetext
Access conditionNo public access
Terms of usehttp://rightsstatements.org/vocab/InC/1.0/
URN:NBNhttps://urn.nsk.hr/urn:nbn:hr:168:931133