Data collection and baselines implementation for phishing detection

• Grant number: 25/26482-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: March 01, 2026
• End date: February 28, 2027
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Anderson de Rezende Rocha
• Grantee: Andrei Quaresma Pinto
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 23/12865-8 - Horus: artificial intelligence techniques to detect and forestall synthetic realities, AP.TEM

Abstract

This undergrad research has the primary objective of preparing a complete (and continuous) dataset of phishing examples through message systems (e.g., SMS, WhatsApp). The idea is to collect data containing examples of attacks and normal communications. The attack examples are based on user annotations and also reports on PhishTank 1 and Google Services 2 .The second objective is to implement existing methods in the literature for phishing detection such as those based on word frequencies, n-grams, and basic recurrent long short-term memory networks (LSTM) and gate-recurrent unit (GRU) deep learning methods.Overall, these research objectives seek to address the growing threat of phishing attacks on messaging platforms by combining traditional natural language processing (NLP) methods with innovative strategies that leverage multimodal fusion and cutting-edge pre-trained language models. (AU)