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

PM2.5 performance analysis under varying imputation strategies for incomplete sensor data.

Rumaisa Chowdhury1, Naveed Ejaz2, Salimur Choudhury1

  • 1School of Computing, Queen's University, Kingston, ON, Canada.

Scientific Reports
|June 30, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Mechanistic Models: Compartment Models in Individual and Population Analysis01:23

Mechanistic Models: Compartment Models in Individual and Population Analysis

Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least squares (OLS)...

You might also read

Related Articles

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

Sort by
Same author

Development and Validation of Machine Learning Models to Optimize Imaging and Referrals for Dizziness in the Emergency Department.

Academic emergency medicine : official journal of the Society for Academic Emergency Medicine·2026
Same author

Enhancing spontaneous recovery after stroke: a randomized controlled trial.

Brain communications·2026
Same author

An interpretable multi-transformer ensemble for text-based movie genre classification.

PeerJ. Computer science·2025
Same author

A Resource-Efficient, High-Dose, Gamified Neurorehabilitation Program for Chronic Stroke at Home: Retrospective Real-World Analysis.

JMIR serious games·2025
Same author

Characterizing the services provided by family physicians in Ontario, Canada: A retrospective study using administrative billing data.

PloS one·2025
Same author

Boundaries of task-specificity: bimanual finger dexterity is reduced in musician's dystonia.

Scientific reports·2024

Forecasting fine particulate matter (PM2.5) is crucial for public health. This study found Kriging imputation combined with LSTM-AM models offers the most accurate PM2.5 predictions, especially with high data availability.

Area of Science:

  • Environmental Science
  • Data Science
  • Public Health

Background:

  • Air pollution, particularly fine particulate matter (PM2.5), is a major global health risk.
  • Accurate PM2.5 forecasting is vital for policy and health impact mitigation.
  • Low-cost sensors (LCS) offer dense, real-time PM2.5 data but suffer from data gaps.

Purpose of the Study:

  • To evaluate the impact of different imputation strategies on PM2.5 forecasting accuracy.
  • To assess the effectiveness of Kriging spatial imputation compared to other methods.
  • To determine the optimal imputation and modeling approach for PM2.5 prediction using LCS data.

Main Methods:

  • Utilized data from eight LCSs in Edmonton, Alberta, including meteorological and geographical features.
Keywords:
Deep learningKriging interpolationLow-cost sensorsMachine learningPM2.5 prediction

Related Experiment Videos

  • Compared Kriging, mean imputation, and k-nearest neighbors against a no-imputation baseline.
  • Evaluated imputation methods' impact on Random Forest, XGBoost, CNN, LSTM, and LSTM with Attention models.
  • Assessed performance across datasets with varying missing data percentages.

    • Main Results:

    • Kriging imputation generally improved PM2.5 forecasting accuracy, with modest differences among imputation methods.
    • Kriging showed particular benefit for deep learning models when data availability was high (90%).
    • The LSTM-AM architecture achieved the best short-horizon forecasting performance (R² = 0.9602, MAE = 1.7815, RMSE = 4.5208).

    Conclusions:

    • Spatial imputation, especially Kriging, can enhance PM2.5 forecasting accuracy.
    • The combination of Kriging imputation and LSTM-AM modeling is optimal for near-term PM2.5 prediction.
    • This approach offers a robust solution for improving air quality monitoring and public health protection.