Abstract: This study proposes a robust ensemble learning framework to significantly reduce false alarms in binary intrusion detection using the CICIDS2017 dataset. The approach integrates mutual ...