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

Air quality prediction using CNN+LSTM-based hybrid deep learning architecture.

Aysenur Gilik1, Arif Selcuk Ogrenci2, Atilla Ozmen2

  • 1Electrical & Electronics Engineering Department, Kadir Has University, Fatih, Istanbul, Turkey. aysenur.gilik@khas.edu.tr.

Environmental Science and Pollution Research International
|September 23, 2021
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

Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data Using Deep Learning: Early Detection of COVID-19 Outbreak in Italy.

IEEE access : practical innovations, open solutions·2021
Same author

ANNSyS: an Analog Neural Network Synthesis System.

Neural networks : the official journal of the International Neural Network Society·2003
Same journal

Application of the incorporation of cigarette butts into building materials: a review.

Environmental science and pollution research international·2026
Same journal

Influence of inoculum-to-substrate ratio on process stability and microbial community structure in anaerobic digestion of human faecal matter.

Environmental science and pollution research international·2026
Same journal

Ecotoxicological and health risk assessing of heavy metals and microplastics in African catfish (Clarias gariepinus) of Burullus Lake, Egypt.

Environmental science and pollution research international·2026
Same journal

Retraction Note: Vitamin D regulates insulin and ameliorates apoptosis and oxidative stress in pancreatic tissues of rats with streptozotocin-induced diabetes.

Environmental science and pollution research international·2026
Same journal

CO<sub>2</sub> removal by applying the adsorption process to biochar from waste materials.

Environmental science and pollution research international·2026
Same journal

Impact of open-air dumping of urban solid waste on soil toxicity and properties in a tropical environment.

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

This study developed a machine learning model for accurate air pollution prediction using sensor data. The model demonstrated improved performance and successful transfer learning between cities with similar environmental conditions.

Area of Science:

  • Environmental Science and Engineering
  • Artificial Intelligence and Machine Learning

Background:

  • Growing concerns about climate change and urban sustainability necessitate accurate air pollution prediction.
  • Modeling complex environmental variables using advanced machine learning techniques is crucial for effective monitoring.

Purpose of the Study:

  • To develop a supervised machine learning model for air pollution prediction using real sensor data.
  • To evaluate the model's performance and its ability to be transferred between different cities.

Main Methods:

  • A hybrid deep learning model combining convolutional neural networks (CNN) and long short-term memory (LSTM) was employed.
  • Both univariate and multivariate models were developed, with the multivariate model incorporating pollutant and meteorological data.
Keywords:
Air pollutionConvolutional neural networkDeep learningLong short-term memoryPredictionTransfer learning

Related Experiment Videos

  • Hyperparameter tuning was performed to optimize model architecture for minimal test error.

    • Main Results:

    • The proposed CNN-LSTM model significantly improved prediction accuracy for particulate matter, ozone, nitrogen oxides, and sulfur dioxide compared to traditional LSTM.
    • The multivariate model, even without meteorological data, yielded the best prediction performance.
    • Transfer learning demonstrated enhanced prediction accuracy when model weights were transferred between geographically and environmentally similar cities (Kocaeli to İstanbul).

    Conclusions:

    • The developed hybrid deep learning model offers a robust approach for accurate air pollution forecasting.
    • The multivariate approach and transfer learning capabilities highlight the model's practical applicability in diverse urban environments.