Ανάλυση δεδομένων τραπεζικού τομέα και τεχνικές μηχανικής μάθησης (Bachelor thesis)

Ρέκα, Μαρτίν


The present research thesis concerns the experimentation with a set of banking data for predicting the smallest possible error in their analysis in order to achieve the best possible loan prediction. Initially, the banking sector is analyzed, including the types of banks and their role in society and the broader economy. Then, the types of loans offered by banks and the ways in which they operate are presented. A significant part of the work, as well as the reason for conducting the research, is addressing the problem of risks that banks face, which are further analyzed. Subsequently, the focus is on risk management, its techniques, and the process involved, as well as the connection between risk management and data analysis. Thus, data analysis is explained, along with some of the most commonly used techniques for conducting it. Then, research articles of similar nature to the thesis research are presented, and a comparison is made among them. Finally, the research process, its results, and their discussion are explained, concluding with the findings and possible future improvements.
Institution and School/Department of submitter: Σχολή Οικονομίας και Διοίκησης - Τμήμα Λογιστικής και Πληροφοριακών Συστημάτων
Subject classification: Τράπεζες και τραπεζικές εργασίες -- Διαχείριση κινδύνου
Δάνεια
Μηχανική μάθηση
Banks and banking -- Risk management
Loans
Machine learning
Keywords: Δεδομένα;Τράπεζες;Δάνεια;Κίνδυνοι;Διαχείριση;Ανάλυση;Έρευνα;Data;Banks;Loans;Risks;Management;Analysis;Research
Description: Πτυχιακή εργασία - Σχολή Οικονομίας και Διοίκησης - Τμήμα Λογιστικής και Πληροφοριακών Συστημάτων, 2023 (α/α 14045)
URI: http://195.251.240.227/jspui/handle/123456789/16819
Appears in Collections:Πτυχιακές Εργασίες

Files in This Item:
File Description SizeFormat 
Reka.pdf2.19 MBAdobe PDFView/Open



 Please use this identifier to cite or link to this item:
http://195.251.240.227/jspui/handle/123456789/16819
  This item is a favorite for 0 people.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.