Scholarlink

<p>Accurate prediction of water-quality indicators remains challenging in small tabular environmental datasets because physicochemical variables can exhibit nonlinear interdependencies and modern deep-learning models are sensitive to hyperparameter configuration. In this study, the supervised regression task is defined as predicting <i>dissolved oxygen</i> <i>(mg/L)</i> from the remaining measured physicochemical variables using a modest public dataset of 200 samples. To address this task, the Automatic Feature Interaction Network (AutoInt) is coupled with the Ninja Optimization Algorithm (NiOA) for wrapper-based hyperparameter optimization under a controlled and reproducible experimental protocol. All evaluations use a fixed train/validation/test split (70%/15%/15%) with leakage-safe preprocessing based only on training-set statistics, and results are aggregated over <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(N=10\)</EquationSource> </InlineEquation> independent runs with controlled initialization seeds and mean ± standard deviation reporting. Baseline AutoInt without metaheuristic optimization achieves a mean squared error of <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(5.812\times 10^{-3}\)</EquationSource> </InlineEquation>, whereas the NiOA-optimized AutoInt configuration reaches <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(1.09\times 10^{-5}\pm 9.64\times 10^{-7}\)</EquationSource> </InlineEquation> under the matched optimization budget of 1500 function evaluations (<InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(P=30\)</EquationSource> </InlineEquation>, <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(T=50\)</EquationSource> </InlineEquation>). These findings indicate that NiOA-guided hyperparameter tuning can substantially improve AutoInt performance within this fixed benchmark setting. However, because the dataset is small, cross-sectional, and lacks explicit spatial or temporal structure, the results should be interpreted as benchmark-specific evidence rather than broad operational validation. Further evaluation on larger, independent, and spatiotemporally diverse water-quality datasets is required before generalizing the fitted model to wider environmental monitoring or decision-support applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Metaheuristic-optimized interaction-aware deep learning with large language model assistance for data-driven water quality prediction

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月21日

查看全文
  • Ebrahim A. Mattar,
  • El-Sayed M. El-Kenawy,
  • Sarah M. Alhammad,
  • Marwa M. Eid

摘要

Accurate prediction of water-quality indicators remains challenging in small tabular environmental datasets because physicochemical variables can exhibit nonlinear interdependencies and modern deep-learning models are sensitive to hyperparameter configuration. In this study, the supervised regression task is defined as predicting dissolved oxygen (mg/L) from the remaining measured physicochemical variables using a modest public dataset of 200 samples. To address this task, the Automatic Feature Interaction Network (AutoInt) is coupled with the Ninja Optimization Algorithm (NiOA) for wrapper-based hyperparameter optimization under a controlled and reproducible experimental protocol. All evaluations use a fixed train/validation/test split (70%/15%/15%) with leakage-safe preprocessing based only on training-set statistics, and results are aggregated over \(N=10\) independent runs with controlled initialization seeds and mean ± standard deviation reporting. Baseline AutoInt without metaheuristic optimization achieves a mean squared error of \(5.812\times 10^{-3}\) , whereas the NiOA-optimized AutoInt configuration reaches \(1.09\times 10^{-5}\pm 9.64\times 10^{-7}\) under the matched optimization budget of 1500 function evaluations ( \(P=30\) , \(T=50\) ). These findings indicate that NiOA-guided hyperparameter tuning can substantially improve AutoInt performance within this fixed benchmark setting. However, because the dataset is small, cross-sectional, and lacks explicit spatial or temporal structure, the results should be interpreted as benchmark-specific evidence rather than broad operational validation. Further evaluation on larger, independent, and spatiotemporally diverse water-quality datasets is required before generalizing the fitted model to wider environmental monitoring or decision-support applications.