DigiNews

Tech Watch by Johan Denoyer

← Back to articles

Detecting scraper bots through scroll behaviour

Quality: 8/10 Relevance: 9/10

Summary

The post explores using burstiness and memory of scroll events to differentiate humans from scraper bots, drawing on prior research. It describes a dataset and a LightGBM model that achieve 73.4% accuracy, and discusses real-world limitations and potential for combining features like mouse movement for improved bot detection.

🚀 Service construit par Johan Denoyer