: 14h45, ngày 15/04/2022 (Thứ Sáu)
: P104 D3
: Seminar Toán rời rạc
: Nguyễn Văn Long
: Cao học Toán Tin, ĐH Bách Khoa Hà Nội
Tóm tắt báo cáo
Malicious calls, i.e., telephony spams and scams, have been a long-standing challenging issue that causes billions of dollars of annual financial loss worldwide. This work presents machine learning-based solutions to detect spam phone numbers . We extensively evaluate various modern machine learning methods including feature based and graph based, then combined some of them by cotrain method, the results also showed that the models are effective for deploying. This presentation will provide a brief review of the methods we used, including Graph Neural Networks