Predikcija prenehavanja igranja na spletnih straneh z igrami na srečo

diplomsko delo

Tara Patricija Bosil (Author), Jure Demšar (Mentor)

Abstract

V današnjem času obstaja velika konkurenca med ponudniki spletnih iger na srečo, zato je zelo pomembno, da ima vsako podjetje jasno strategijo, kako pridobiti oziroma predvsem obdržati svoje stranke. Glede na to, da je za pridobitev nove stranke potreben bistveno večji vložek kot za obdržanje že obstoječe stranke, je cilj diplomske naloge, da s pomočjo podatkov podjetja Oryx gaming d.o.o poskušamo napovedati, ali bo igralec zapustil spletno igralnico. Problem prenehavanja igranja na spletnih igralnicah smo rešili s pomočjo metod napovedovanja prenehanja uporabe storitev (angl. \textit{churn prediction}). Za napoved smo uporabili dve statistični metodi, logistično regresijo in Bayesovo logistično regresijo. Glede na dobljene rezultate smo ugotovili, da je uspešnost omenjenih metod pri napovedovanju zelo podobna. Končen klasifikacijski model je pri napovedovanju, ali bo igralec zapustil spletno igralnico ali ne, natančen v 73\%. Na podlagi rezultatov lahko tako spletna igralnica sprejme različne ukrepe, ki bodo pripomogli k zmanjšanju odliva trenutnih igralcev.

Keywords

napovedovanje;logistična regresija;Bayesova logistična regresija;računalništvo;računalništvo in informatika;računalništvo in matematika;interdisciplinarni študij;univerzitetni študij;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2019
Typology:	2.11 - Undergraduate Thesis
Organization:	UL FRI - Faculty of Computer and Information Science
Publisher:	[T. P. Bosil]
UDC:	004(043.2)
COBISS:	1538315715
Views:	1312
Downloads:	293
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Churn prediction in online casinos
Secondary abstract:	Nowadays there is a lot of competitiveness between online casinos. Because of that, every company must have a clear strategy how to attract or even more importantly how to keep their customers. Given that the acquirement of a new customer consumes much more resources than just keeping an already existing one. The aim of this bachelor's degree is to facilitate churn prediction methods on data about the bahavior of online gamblers from Oryx gaming d.o.o. and predict whether a certain player intends to leave an online casino. We used two statistical methods to execute this prediction, logistic regression and Bayesian logistic regression. In our case the results from both methods are comparable. Our final model's accuracy at predicting if a customer will stop gambling or not is 73\%. According to the results an online casino can take different actions, which will help to keep existing players active.
Secondary keywords:	churn prediction;logistic regression;Bayesian logistic regression;computer science;computer and information science;computer science and mathematics;interdisciplinary studies;diploma;
Type (COBISS):	Bachelor thesis/paper
Study programme:	1000407
Embargo end date (OpenAIRE):	1970-01-01
Thesis comment:	Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages:	34 str.
ID:	11214234