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 Videos

Deep learning architecture for air quality predictions.

Xiang Li1,2, Ling Peng3, Yuan Hu1,2

  • 1Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing, 100094, China.

Environmental Science and Pollution Research International
|October 14, 2016
PubMed
Summary
This summary is machine-generated.

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

ITLN1 Improves Endothelial Dysfunction in Hypertensive Mice via Wnt5b-JNK Signaling.

Hypertension (Dallas, Tex. : 1979)·2025
Same author

Effectiveness, ethics, and sustainability of nudge-based interventions for self-monitoring in patients with hypertension and type 2 diabetes: A systematic review.

Health psychology : official journal of the Division of Health Psychology, American Psychological Association·2025
Same author

Asiatic acid alleviates dexamethasone-induced muscle atrophy through regulating the Sirt1/PGC-1α/FOXO3 pathway.

Histology and histopathology·2025
Same author

Fabrication and characterization of zein/carboxymethyl chitosan nanoparticles for co-encapsulation of curcumin and resveratrol.

Frontiers in nutrition·2025
Same author

[Progress on the application of system dynamics model in the field of health management].

Zhejiang da xue xue bao. Yi xue ban = Journal of Zhejiang University. Medical sciences·2025
Same author

20-Deoxyingenol attenuated doxorubicin-induced cardiotoxicity by promoting autolysosome degradation through the UCHL3-TFEB pathway.

Phytomedicine : international journal of phytotherapy and phytopharmacology·2025
Same journal

Mineralogical characterization of tobacco stalk ash.

Environmental science and pollution research international·2026
Same journal

Occurrence, distribution and risk assessment of pharmaceuticals and personal care products in Djibouti aquifer.

Environmental science and pollution research international·2026
Same journal

Anthropogenic microparticles and mercury co-occurrence in blue sharks from the Tropical Eastern Pacific.

Environmental science and pollution research international·2026
Same journal

IoT-enabled FMIND pipeline with chemical validation for microplastic contamination risk assessment in bottled water under varying storage conditions.

Environmental science and pollution research international·2026
Same journal

Retraction Note: Toward a sustainable environment: nexus between geothermal energy growth and land use change in EU economies.

Environmental science and pollution research international·2026
Same journal

Retraction Note: COVID-19: pathogenesis, advances in treatment and vaccine development and environmental impact-an updated review.

Environmental science and pollution research international·2026
See all related articles

This study introduces a novel deep learning method for accurate air quality prediction, outperforming traditional models. The spatiotemporal deep learning approach enhances forecasting for better environmental and health management.

Area of Science:

  • Environmental Science
  • Computer Science
  • Data Science

Background:

  • Urbanization and industrialization drive significant air pollution in developing nations.
  • Existing shallow models for air quality prediction yield unsatisfactory results.
  • Growing concerns exist regarding air pollution's impact on public health and sustainable development.

Purpose of the Study:

  • To propose a novel spatiotemporal deep learning (STDL) method for air quality prediction.
  • To address limitations of current prediction models by incorporating spatial and temporal correlations.
  • To improve the accuracy and reliability of air quality forecasting.

Main Methods:

  • Utilized a stacked autoencoder (SAE) for extracting inherent air quality features.
Keywords:
Air quality predictionBP algorithmDeep learningLayer-wise pre-trainingSpatiotemporal featuresStacked autoencoder (SAE)

Related Experiment Videos

  • Employed a greedy layer-wise training approach for the SAE model.
  • Developed a novel spatiotemporal deep learning architecture.

    • Main Results:

    • The proposed STDL method demonstrated superior performance compared to traditional models.
    • The model achieved simultaneous air quality prediction across all stations.
    • The method exhibited temporal stability across different seasons.

    Conclusions:

    • The developed STDL-based air quality prediction method offers a significant advancement over existing techniques.
    • This approach provides more accurate and stable air quality forecasts.
    • The findings support the use of deep learning for effective environmental monitoring and management.