An innovative hybrid model based on outlier detection and correction algorithm and heuristic intelligent optimization...

Area of Science:

• Environmental Science
• Data Science
• Computational Science

Background:

• Accurate air pollution forecasting is crucial for emission reduction and public health advisories.
• Existing forecasting models often neglect outlier detection and parameter optimization, impacting reliability.

Purpose of the Study:

• To develop a hybrid model that addresses limitations in current air pollution forecasting methods.
• To enhance prediction accuracy through outlier management and intelligent parameter optimization.

Main Methods:

• Data preprocessing techniques for outlier detection and correction in time series.
• Optimization of Extreme Learning Machine (ELM) parameters using a heuristic intelligent algorithm.
• Hybrid model development integrating outlier handling and optimized ELM for subseries forecasting.

Main Results:

• The proposed hybrid model demonstrates superior prediction accuracy compared to existing models.
• Outlier detection and correction significantly improve the reliability of air pollution forecasts.
• Optimized model parameters enhance the performance of the Extreme Learning Machine.

Conclusions:

• The hybrid model offers a robust and accurate approach to air pollution forecasting.
• Emphasizes the critical role of data preprocessing and parameter optimization in time series analysis.
• Provides a feasible method for effective air quality management and emission control strategies.