This research introduces SENIR (Software-specific Named Entity Recognition, Intent Detection, and Resolution Classification) to enhance question clarity in software engineering chatrooms.

• Addresses imprecise developer questions that hinder effective communication
• Leverages LLMs to identify software-specific entities missed by traditional NLP approaches
• Creates a specialized approach for understanding technical conversations in development contexts
• Aims to improve resolution of developer queries through better entity recognition

For engineering teams, this work offers potential improvements in knowledge sharing, troubleshooting efficiency, and developer productivity by enhancing the quality of technical communication.

Towards Refining Developer Questions using LLM-Based Named Entity Recognition for Developer Chatroom Conversations