IMMENSE: AI Method to Detect Malicious Social Media Users
A novel machine learning technique called IMMENSE has been developed to detect harmful users on social media who share discriminatory, hateful, and violent messages. This approach, outlined in a paper available on arXiv (arXiv:2608.05259), employs a hybrid classification model that combines three dimensions: the semantics of the content shared by users, their social connections, and their geographical data. By utilizing this comprehensive strategy, the classification accuracy is expected to surpass that of traditional text-only methods. The need for such systems is highlighted by the vastness of current social networks and the susceptibility of younger users to damaging content, which can lead to psychological and social issues. This method aims to support Law Enforcement Agencies (LEAs) in pinpointing and managing users who spread harmful material.
Key facts
- IMMENSE is a machine learning-based method for detecting malicious social network users.
- It integrates three perspectives: content semantics, social relationships, and spatial information.
- The method is described in a paper on arXiv with identifier 2608.05259.
- The paper is a cross-announcement type.
- The method targets users who propagate discriminatory, hateful, and violent content.
- Young users are particularly vulnerable to exposure to such content.
- The system is intended to aid Law Enforcement Agencies (LEAs).
- The approach enhances classification performance beyond text-only analysis.
Entities
Institutions
- arXiv