Automated Alignment between Elicitation Interviews and Requirements
Researchers have developed a method to automatically align interview transcripts with software requirements, represented as user stories. They introduced two metrics: "requirements faithfulness" and "interview coverage." Experiments using large language models demonstrated an 0.86 macro-F1 score on labeled data, suggesting LLMs can effectively evaluate these metrics. This work aims to streamline the process of tracing conversational artifacts to requirements and potentially generate requirements from conversations. AI
IMPACT Automates a key step in software development, potentially speeding up requirement gathering and reducing errors.