TY - GEN
T1 - AdaptPhishSysNet
T2 - 3rd World Conference on Information Systems for Business Management, ISBM 2024
AU - Manoj, R.
AU - Joshi, Sandeep
AU - Aravinthakshan, A. S.
AU - Tripti, C.
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
PY - 2025
Y1 - 2025
N2 - The decentralized digital era demands applications that keep your wallet credentials safe. As more people embrace digital currencies and online financial transactions, the need for robust security measures becomes increasingly important. Transactions on a blockchain network are relatively safe because of their decentralized nature. Still, the website in which the credentials are entered becomes a point of vulnerability or weakness to Phishing attacks targeting wallet applications and smart contracts, which can be particularly harmful, as they may result in loss of funds or compromising sensitive information. These attacks aim to trick users into revealing sensitive information, such as usernames, passwords, or private keys, by impersonating a trustworthy entity. This study presents a system for identifying phishing URLs, by utilizing a type of 1D ResNet that learns incrementally, combined with a continuously updating blacklists system to stay current with the evolving trends in malicious URL configurations. The proposed scheme AdaptPhishSysNet is compared with various machine learning models and other URL phishing detection methods, and the results show that the scheme can classify the phishing attacks in the long run without a high toll on the model’s ability to generalize.
AB - The decentralized digital era demands applications that keep your wallet credentials safe. As more people embrace digital currencies and online financial transactions, the need for robust security measures becomes increasingly important. Transactions on a blockchain network are relatively safe because of their decentralized nature. Still, the website in which the credentials are entered becomes a point of vulnerability or weakness to Phishing attacks targeting wallet applications and smart contracts, which can be particularly harmful, as they may result in loss of funds or compromising sensitive information. These attacks aim to trick users into revealing sensitive information, such as usernames, passwords, or private keys, by impersonating a trustworthy entity. This study presents a system for identifying phishing URLs, by utilizing a type of 1D ResNet that learns incrementally, combined with a continuously updating blacklists system to stay current with the evolving trends in malicious URL configurations. The proposed scheme AdaptPhishSysNet is compared with various machine learning models and other URL phishing detection methods, and the results show that the scheme can classify the phishing attacks in the long run without a high toll on the model’s ability to generalize.
UR - https://www.scopus.com/pages/publications/105007992469
UR - https://www.scopus.com/pages/publications/105007992469#tab=citedBy
U2 - 10.1007/978-981-96-1747-0_32
DO - 10.1007/978-981-96-1747-0_32
M3 - Conference contribution
AN - SCOPUS:105007992469
SN - 9789819617463
T3 - Lecture Notes in Networks and Systems
SP - 389
EP - 399
BT - Information Systems for Intelligent Systems - Proceedings of ISBM 2024
A2 - Iglesias, Andres
A2 - Shin, Jungpil
A2 - Patel, Bharat
A2 - Joshi, Amit
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 12 September 2024 through 13 September 2024
ER -