Internet and Rumors - Identify It Faster

Early rumor detection is difficult because signals are sparse at the beginning of an event, when intervention matters most.

The proposed approach combines dual CNN pipelines and certainty-factor-based activation to improve feature extraction under low-data conditions.

This work demonstrates how architecture decisions can target practical constraints such as limited early evidence and response urgency.