Primjena dubinske analize podataka o korisnicima telekomunikacijskih usluga

scientific master's theses

• Title (english): Data mining of user data in telecommunication services
• Author: Rafael Buždon
• Mentor: Zoran Skočir (mentor)
• Committee member: Ignac Lovrek (predsjednik povjerenstva)
• Committee member: Zoran Skočir (član povjerenstva)
• Committee member: Dragan Gamberger (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Telecommunications); Zagreb
• Defense date and country: 2010-11-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

U okviru ovog magistarskog rada prikazana je primjena dubinske analize podataka kao jedne od discipline poslovne inteligencije u okruženju telekomunikacijskog tržišta, s naglaskom na pokretnu telefoniju, kroz konkretan primjer predviđanja smanjenja intenziteta prometa korisnika. Tržište telekomunikacijskih usluga danas, osobito pokretna telefonija, je veoma dinamično i konkurentno,. Penetracija tržišta u mnogim zemljama, osobito razvijenih zemalja, je prešla 100% čime je primorao davatelje usluga na promjenu načina poslovanja iz poslovanja usmjerenog prema pridobivanju novog korisnika prema poslovanju usmjerenog prema klijentu sa strategijom zadržavanja korisnika. U novije doba dubinska analiza podataka postala je jedan od najvažnijih načina otkrivanja znanja u velikim bazama podataka u svrhu provedbe tih novih načina poslovanja. U radu je prikazan proces dubinske analize podataka na jednom segmentu pre-paid korisnika operatora pokretne telefonije u svrhu predviđanja smanjenja intenziteta prometa: od procesa prikupljanja podataka iz korporativnog skladišta podataka, preko analize i transformacije podataka, izgradnje prediktivnih modela korištenjem SPSS PASW Modeler alata, odabir modela s najboljim rezultatima, te konačno procesa implementacije najboljih modela u poslovni proces. Prediktivni modeli napravljeni su korištenjem triju metoda raspoznavanja uzorka: regresija, stabla odluke i neuronske mreže. Također je opisan i način mjerenje kvalitete rezultata prediktivnih modela na osnovu kojih su odabrani najbolji. Na kraju rada dani su rezultati primjene modela u stvarnom svijetu usporedbom odziva korisnika odabranih modelom i ostalih korisnika dobivenih slučajnim odabirom – kontrolna skupina.

Abstract (english)

In this paper we demonstrate the application of data mining as a discipline of business intelligence in telecommunications, with emphasis on mobile telecommunications, through a specific example of prediction client traffic decrease. Telecommunication market today is very dynamic and competitive, primely in mobile telecommunications. The market penetration in many countries, especially in the developed countries, has reached 100% and thus forcing service providers to change its way of doing business from acquisition oriented to retention oriented. Recently, data mining has become one of the most important process of knowledge discovering in huge databases for purpose of supporting this new retention oriented companies. In this paper we describe data mining process of mobile operator on one segment of pre-paid customers for the purpose of prediction customer traffic decrease: beginning at data collection from the corporative data warehouse, through data analyze and transformation, building prediction models using the SPASS PASW Modeler tool, assessing models and finally implementation of chosen models in the production. Prediction models have been built by using of three pattern recognitions methods: regression, decision tree and neural networks. Additionally, model assessing has been described on which basis the best model has been chosen. At the end of this paper we assess results of real life applications of models by comparing the number of responders from clients chosen by prediction models and the number of responders from randomly chosen clients – control group.

• Keywords: telekomunikacije; zadržavanje korisnika; dubinska analiza podataka; prediktivni model; smanjenje intenziteta prometa korisnika
• Keywords (english): telecommunications; client retention; data mining; prediction model; client traffic decrease
• Language: croatian
• URN:NBN: urn:nbn:hr:168:510819
• Project: Number: 036-0362027-1638; Title: Umrežena ekonomija; Leader: Zoran Skočir; Jurisdiction: Croatia; Funder: MZOS; Funding stream: ZP
• Study programme: Title: Postgraduate master programme in electrical engineering; Study programme type: university; Study level: postgraduate; Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
• Catalog URL: http://lib.fer.hr/cgi-bin/koha/opac-detail.pl?biblionumber=34331
• Type of resource: Text
• Extent: 115 str ; 30 cm
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-08-19 08:08:35