About this document
Unsupervised Twitter Bot Detection Method by Akeem Cruz is a document available to read on EtoBox.
The document presents an unsupervised method for detecting Twitter social bots using deep contrastive graph clustering (BotDCGC), which addresses the limitations of traditional supervised methods that struggle with evolving bot behaviors. By utilizing a graph attentional encoder and contrastive learning techniques, the model effectively learns user node embeddings and performs clustering without relying on labeled data. Experimental results demonstrate that BotDCGC outperforms existing state-of-the-art meth
- Author
- Akeem Cruz
- Language
- EN