GN5-2 WP6 Incubator Project Closure Report: LogsLLM: Automated Log Analysis and Management based on Large Language Models (LLMs)

Key Points

• The aim is to deliver an automated log analysis and management solution using large language models for better incident summarization.
• Developed a standardisation pipeline for normalising heterogeneous logs into a unified schema.
• Created an analysis component that correlates alerts across systems to generate root-cause-oriented summaries.
• Validated the solution in four operational scenarios using multiple LLMs, including ChatGPT-4.1 mini and Llama 3.1 8B.
• The system produced accurate incident summaries suitable for Network Operations Center (NOC) workflows.
• It facilitated faster triage and significantly reduced alert fatigue among operations teams.