I’m proud to announce that our latest article WikiPhish: A Diverse Wikipedia-Based Dataset for Phishing Website Detection has been accepted at CODASPY 2024 (ACM Conference on Data and Application Security and Privacy). TL;DR: Phishing poses a persistent threat, demanding advanced detection systems. Supervised machine learning is widely employed to automate detection, reliant on extensive annotated data. The introduction of WikiPhish dataset, comprising 110,606 webpages (from Wikipedia, OpenPhish and PhishTank), addresses this need for diverse and robust data, enhancing phishing detection model development.

IntroductionIn today’s digital world, APIs serve as the core of modern applications, allowing seamless data exchange and interaction. However, even the most robust and scalable API servers can succumb to the humongous number of incoming requests. Picture this scenario: your finely crafted API server, designed to handle thousands of requests per second, suddenly experiences an unprecedented spike in traffic. Within moments, your server starts to struggle under the weight of the overwhelming load, its response times slowing to a crawl, and in the worst-case scenario, a complete outage occurs.

I’m proud to announce that our latest article pre-print has been published on arxiv! It has also been submitted to a journal, and hopefully be published in the next few months. TL;DR: Cybercrime, a major cause of financial losses and intellectual property theft, often begins with email, the primary means of corporate communication. Spear phishing or Business Email Compromise are rare, damaging email threats. However, literature on targeted attacks is inconsistent, hindering security vendors and researchers from mitigating them.

This is my first post, yay! Nothing much to say apart from that! Feel free to check my publications and about pages if you want to know more about me.