Cyber-attacks have become a significant challenge to organizations worldwide. MITRE ATT&CK framework is a widely adopted methodology for identifying, categorizing, and describing different types of cyber-attacks. However, detecting and preventing these attacks remain a significant challenge for cybersecurity researchers and practitioners. In this study, we assessed the ability of the BiLSTM Classifier to detect attacks in different stages of MITRE ATT&CK frameworks. Our results show that the BiLSTM classifier can achieve high accuracy in detecting attacks across various stages of the framework, including reconnaissance, initial access, execution, persistence, and exfiltration.