Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 23, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

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

Ebrahim A Mattar1, El-Sayed M El-Kenawy2,3, Sarah M Alhammad4

  • 1College of Engineering, University of Bahrain, Zallaq, Bahrain.

Scientific Reports
|June 21, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Geotechnical challenges of urban expansion in Mila Town (NE Algeria): an integrated Engineering Ground Model (EGM) approach.

Scientific reports·2026
Same author

Air quality prediction via hybrid BiGRU-MLP using binary ant colony optimization and firefly algorithm for hyperparameter tuning.

Scientific reports·2026
Same author

Human-inspired hyperparameter optimization for long-horizon forecasting of freshwater and desalination per-capita dynamics.

Scientific reports·2026
Same author

An optimization-driven hierarchical deep learning approach using the Gray Langurs algorithm for data-driven seismic activity prediction.

Scientific reports·2026
Same author

Deposit characterizations and engineering-geotechical modeling for sustainable urbanisation in the Mila basin (NE Algeria).

Scientific reports·2026
Same author

Structured vital sign prediction in hospital environments via an Al-Biruni earth radius optimization-driven unified metaheuristic framework.

Scientific reports·2026

Predicting water quality dissolved oxygen is challenging with small datasets. Combining Automatic Feature Interaction Network (AutoInt) with Ninja Optimization Algorithm (NiOA) significantly improved prediction accuracy by optimizing hyperparameters.

Area of Science:

  • Environmental Science
  • Data Science
  • Machine Learning

Background:

  • Accurate water-quality prediction is crucial but difficult with limited environmental data.
  • Physicochemical variables often have complex nonlinear relationships.
  • Deep learning models require careful hyperparameter tuning, especially on small datasets.

Purpose of the Study:

  • To predict dissolved oxygen (mg/L) using physicochemical variables from a small tabular dataset.
  • To evaluate the effectiveness of coupling the Automatic Feature Interaction Network (AutoInt) with the Ninja Optimization Algorithm (NiOA) for hyperparameter optimization.
  • To establish a reproducible benchmark for water-quality prediction models.

Main Methods:

  • A supervised regression task was defined to predict dissolved oxygen.
Keywords:
Environmental data modelingHyperparameter tuningMetaheuristic optimizationReproducible benchmarkingTabular deep learningWater quality prediction

Related Experiment Videos

Last Updated: Jun 23, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

  • The Automatic Feature Interaction Network (AutoInt) model was employed.
  • Wrapper-based hyperparameter optimization was performed using the Ninja Optimization Algorithm (NiOA).
  • A fixed 70%/15%/15% train/validation/test split and leakage-safe preprocessing were used.
  • Results were aggregated over multiple independent runs with controlled seeds.

    • Main Results:

    • Baseline AutoInt achieved a mean squared error of [Formula: see text].
    • NiOA-optimized AutoInt reached a mean squared error of [Formula: see text] within 1500 function evaluations.
    • NiOA-guided hyperparameter tuning substantially improved AutoInt performance on this benchmark.

    Conclusions:

    • NiOA-guided hyperparameter optimization significantly enhances AutoInt performance for dissolved oxygen prediction in small datasets.
    • The study provides benchmark-specific evidence for the efficacy of the proposed method.
    • Further validation on diverse, larger datasets is necessary for broader applicability in environmental monitoring.