Researchers have improved the classification of software requirements by applying the SMOTE-Tomek preprocessing technique to the PROMISE dataset. This method effectively addresses class imbalance within the dataset, which contains 969 requirements categorized as functional or non-functional. The approach led to a significant increase in classification accuracy, with logistic regression achieving 76.16% compared to a baseline of 58.31%, demonstrating the utility of machine learning for scalable and interpretable solutions in requirements engineering. AI
IMPACT Enhances machine learning model performance in software engineering by addressing data imbalance.
RANK_REASON This is a research paper detailing a new method for improving classification accuracy on a specific dataset. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
