Семинар ММИТ "Catch me if you can: A participant-level rumor detection framework via fine-grained user representation learning"

15 ноября 2024 г. в 13:00  в рамках семинара "Математические модели информационных технологий" департамента анализа данных и искусственного интеллекта и МЛ "Интеллектуальных систем и структурного анализа" под руководством С.О. Кузнецова состоится в формате онлайн конференции выступление Рашида Бензадидооста.

Тема: Catch me if you can: A participant-level rumor detection framework via fine-grained user representation learning

 Докладчик: Rashid Behzadidoost, Ph.D. Candidate in computer science at University of Tabriz

Аннотация:

Researchers have made significant advances in automatic rumor detection. Traditional models focus on feature engineering, which requires a lot of human effort and is difficult to generalize.
While deep learning models provide improvements, they often have limitations in capturing the detailed dynamics of rumor spread, including user influence and interaction patterns. A participant-level framework using fine-grained user representations better addresses these complexities by modeling factors like user influence, susceptibility, and temporal patterns in rumor propagation, leading to improved accuracy gains in real-world datasets. In this presentation, we discuss the user representation learning model for rumor detection and show key results and insights.

 Ссылка на видеовстречу: https://telemost.yandex.ru/j/8534690121