Churn prediction at insurance companies by intelligent data analytics

OData support
Supervisor:
Dr. Szűcs Gábor
Department of Telecommunications and Media Informatics

Nowadays it's getting more and more important for customer-focused companies to keep their partners who they have already convinced. However as a result of the continuous competition in the market it is a difficult task, which calls for more resources. We call it a drop-out if someone chooses to leave the company.

In my thesis, one of my goals is to recognize and find a pattern for drop-outs and acquire experiences from them, while introducing the terminologies and products of the insurance sphere sector and basic data mining methods.

In my thesis, I first studied the world of data analysis, detailing the sequence of the data mining methods, following this I introduce various data mining algorithms and their success criteria. In the second stage, I reported on the main organizations and key products of the insurance sector, presenting the professional background of the problem at hand. Following this stands the solution of my task, where I’m detailing the analysis of the acquired data set, the built prediction models and also the evaluation of the results.

Downloads

Please sign in to download the files of this thesis.