Fraud Detection in Banking
Ha, Anh (2023)
2023
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2023101927763
https://urn.fi/URN:NBN:fi:amk-2023101927763
Tiivistelmä
The thesis examines the process of fraud detection from financial banking’s perspective. Along with technology advancement, the number of committed frauds has been on the rise and shows no sign of stopping. Consequently, banks are under enormous pressure to protect their client’s assets and their own reputations.
In the fight against fraud, many banks are still reliable on old-school data analysis techniques with the prime example being rules-based system. Such systems show clear weakness against high volume of fraud as it needs constant feedback loops, leading to inefficiency as time is scare in detecting and preventing fraud. Moreover, ‘a victim zero’ is always needed, pushing banks into a passive position where they are always one step behind fraudsters.
The thesis aims to provide a more modern approach that help banks to gain better leverage in their battles. By utilizing machine learning, banks can automate the detection process. Adopting machine learning approach will give banks an upper hand in this fight as it allows banks to accurately predict and detect both previous acknowledged and unrecognized fraud modus.
In the fight against fraud, many banks are still reliable on old-school data analysis techniques with the prime example being rules-based system. Such systems show clear weakness against high volume of fraud as it needs constant feedback loops, leading to inefficiency as time is scare in detecting and preventing fraud. Moreover, ‘a victim zero’ is always needed, pushing banks into a passive position where they are always one step behind fraudsters.
The thesis aims to provide a more modern approach that help banks to gain better leverage in their battles. By utilizing machine learning, banks can automate the detection process. Adopting machine learning approach will give banks an upper hand in this fight as it allows banks to accurately predict and detect both previous acknowledged and unrecognized fraud modus.