Please use this identifier to cite or link to this item: http://hdl.handle.net/20.500.12188/17586
Title: Анализа на големи податоци во банкарскиот сектор
Other Titles: Analysis of Big Data in banking sector
Authors: Доко, Фисник
Keywords: credit risk, credit registry, natural language processing, named entity recognition, sentiment analysis
Issue Date: 2021
Publisher: ФИНКИ, УКИМ, Скопје
Source: Доко, Фисник (2021). Анализа на големи податоци во банкарскиот сектор. Докторска дисертација. Скопје: ФИНКИ, УКИМ.
Abstract: We live in a world where exponential growth of data is transforming the way that companies work by providing new knowledge and insights that are valuable for making informed decisions. Big Data represents potential treasure for companies which will know to successfully manage and utilize it. The financial sector is guided by data in its decisions, also managing the various risks in day-to-day operations. The usage of Big Data and Data Science in banks provides many benefits, enabling them to better manage various risks and gain a competitive advantage. The most common challenge in financial institutions is the credit risk assessment, and its correct prediction is crucial for them. There is vast amount of research on credit risk but all predictions are based on the use of data sets from commercial banks, and to my knowledge there is no research that uses data from the Credit Registry database which are available only in central banks. Financial news represents an important information that influence and predict various financial instruments. They also can be used for analysis of companies by analyzing public textual content for them. The uniqueness of the dissertation is that so far there is no research that predicts credit risk using the Credit Registry database and using the additional knowledge from financial news in Macedonian and Albanian language. The doctoral dissertation proposes a platform for predicting credit risk. The platform integrates credit risk prediction for a client or company using data from the Credit Registry. In the platform there is also integrated analysis of financial news using Named Entity Recognition and Sentiment Analysis. These two separate models are created using financial news in Macedonian and Albanian language. The platform enables better credit risk management with the help of additional opinion from the central bank. Credit risk analysis is helped additionally by financial news that analyzes the company in which the client is employed or the company that is applying for a loan. The platform is designed for economists and brokers, providing them the unique benefits for dealing with credit risk.
Description: Докторска дисертација одбранета во 2021 година на Факултетот за информатички науки и компјутерско инженерство во Скопје, под менторство на проф. д–р Игор Мишковски.
URI: http://hdl.handle.net/20.500.12188/17586
Appears in Collections:UKIM 02: Dissertations from the Doctoral School / Дисертации од Докторската школа

Files in This Item:
File Description SizeFormat 
S-FisnikDoko2021.pdf4.02 MBAdobe PDFView/Open
Show full item record

Page view(s)

107
checked on May 27, 2024

Download(s)

128
checked on May 27, 2024

Google ScholarTM

Check


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