Researchers have developed a new method to detect duplicate steps in Behaviour-Driven Development (BDD) using Gherkin, a common language for specifying software behavior. This approach addresses the significant maintenance burden caused by redundant step definitions in large BDD suites. The system utilizes a combination of exact hashing, Levenshtein distance, and sentence transformer embeddings to identify both exact and semantically similar duplicates, and it is validated against a newly created benchmark of over one million steps. AI
IMPACT Introduces a novel approach to reduce maintenance costs in software development by identifying duplicate code patterns.
RANK_REASON This is a research paper detailing a new method and benchmark for detecting duplicates in software development artifacts.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
