WEPHISH PHISHING WEBSITE DETECTOR BASED ON DEEP LEARNING TECHNOLOGY
BOOTH NO : H3
Category : Information Communication Technology & Multimedia
Phishing has become an increasing concern and captured the attention of end-users as well as security experts. Existing phishing detection solutions still suffer from the deficiency in performance accuracy despite decades of development and improvement. Motivated to solve this problem, many researchers in the cyber security domain have shifted their attention to phishing detection that capitalizes on machine learning techniques. Deep learning has emerged has a branch of machine learning that becomes a promising solution for phishing detection in recent years. However, most of the current deep learning models were trained on poor-quality datasets, used non-representative features for extraction, and relied on single deep learning algorithm for classification. As a result, this research proposed a phishing website detector, named WePhish, based on ensemble deep learning model and multi-dimensional features extracted from a high-quality dataset, to improve the detection accuracy.
In recent years, phishing detection solutions based on machine learning have received tremendous attention from researchers across the globe. However, most of the existing machine learning-based phishing detection models suffer from limited detection accuracy due to being trained on poor-quality datasets, using non-representative features for extraction, and relying only on a single algorithm for classifying phishing websites. In addition, traditional machine learning approaches are unable to handle a large amount of data in the big data era. Moreover, current detection toolkits using conventional machine learning techniques also fail to detect newly evolving phishing patterns. Phishing website detection based on deep learning is still an emerging research area that has not yet been fully explored. Consequently, there is an urgent need for a robust, scalable, and flexible phishing detection model with high performance accuracy to effectively identify phishing and legitimate websites.
90% of data breaches are caused by phishing.
76% of organizations are victims of phishing attacks.
Phishing-as-a-Service (PhaaS) is currently on the rise.
Phishing causes huge financial damage to organizations and loss of trust in individuals.
A phishing website detector is developed to detect malicious websites effectively and efficiently.
A business model will be prepared, so that the idea of this innovation can be proposed to targeted organizations or companies in Malaysia.
Easy to use. High detection accuracy.Produce quick result.Avoid financial loss and damage
Malaysia government ministry and agency. Other researchers from other universities that work on the same project.
Malaysia (Cyber Security Malaysia, National Cyber Security Agency).
Regional: Southeast Asia
Asia
ALI BIN SELAMAT
JABATAN TIMBALAN NAIB CANSELOR (HAL EHWAL PELAJAR & ALUMNI)
ALI BIN SELAMAT
JABATAN TIMBALAN NAIB CANSELOR (HAL EHWAL PELAJAR & ALUMNI)
DO NGUYET QUANG
MALAYSIA-JAPAN INTERNATIONAL INSTITUTE OF TECHNOLOGY