Machine learning should be judged by how well it predicts, and computational resources are not accounted for in predictive accuracy. Given the growing emphasis on energy consumption and resource efficiency, decision-supporting frameworks should go beyond accuracy. This study presents an energy-based benchmarking approach for supervised learning models. Ten classical algorithms were evaluated on three textual and tabular datasets. The energy consumption of preprocessing, training, and inference was monitored with Intel RAPL via pyRAPL along with the runtime, peak memory usage, and predictive performance statistics (accuracy, precision, recall, F1-score, and AUC). Experiments were conducted in a controlled CPU-based environment to ensure comparability. The computational role of this feature is found to be appreciably diverse. Results show that Random Forest achieved the highest overall balance between predictive performance and efficiency (CI = 0.950, PPI = 0.907), while Logistic Regression provided a competitive trade-off (CI = 0.905, EI = 0.998). Gaussian Naïve Bayes was the most energy-efficient model with a mean energy consumption of 127 J, whereas Support Vector Classifier (SVC) incurred the highest computational cost, consuming 45,758 J and requiring 3925 s on average. The Pareto analysis identified Random Forest, Logistic Regression, Passive Aggressive, and Decision Tree as non-dominated solutions. These findings demonstrate that accuracy alone can be misleading for model evaluation and that integrating energy, runtime, and memory metrics enables more sustainable and resource-aware machine learning model selection. The proposed framework provides practical guidance for Green AI, Tiny Machine Learning (TinyML), edge computing, and other resource-constrained deployment environments.
The process of authoring a Software Requirements Specification (SRS) document is a resource-intensive task in software development that requires coordination among multiple stakeholders and is often time-consuming, costly, and prone to human error. Latest advancements in artificial intelligence have enabled the generation of specification documents using Large Language Models (LLMs). However, such approaches still depend on manual prompt engineering and prompt optimization to extract relevant knowledge and do not consistently ensure structural coherence, completeness, and reliability. This paper presents a systematic approach to SRS generation in which input requirements from stakeholders are classified into semantically meaningful topics, followed by the construction of an initial skeleton document based on these topics. The document is then incrementally expanded using reinforcement learning to improve consistency, completeness, and coverage. The proposed approach also improves the dependability of the output by reducing hallucinations that may arise from the unstructured, raw nature of user inputs. The experimental evaluation of the proposed framework increases topic classification accuracy from 0.40–0.70 to 0.75–0.95 across six requirement topics, thereby improving document structure and generation quality. Compared with a ChatGPT Model 5.2 baseline, the framework achieved significant improvements in key text-generation metrics, including a 19.8% increase in ROUGE-L and an 11.8% increase in METEOR, while maintaining contextual relevance with an average semantic cross-similarity score of 0.615. These results indicate that the proposed method can produce SRS documents that are contextually reliable and structurally coherent while requiring less manual prompts such as requirements from analysts or stakeholders.
Muhammad Ali Akhtar, Raheela Asif· Information· 0 citations