We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
AQIPred: A Hybrid Model for High Precision Time Specific Forecasting of Air Quality Index with Cluster Analysis.
- Authors
Yasmin, Farhana; Hassan, Md. Mehedi; Hasan, Mahade; Zaman, Sadika; Angon, Jarif Huda; Bairagi, Anupam Kumar; Changchun, Yang
- Abstract
The discipline of forecasting and prediction is witnessing a surge in the application of these techniques as a direct result of the strong empirical performance that approaches based on machine learning (ML) have shown over the past few years. Especially to predict wind direction, air and water quality, and flooding. In the context of doing this research, an MLP-LSTM Hybrid Model was developed to be able to generate predictions of this nature. An investigation into the Beijing Multi-Site Air-Quality Data Set was carried out in the context of an experiment. In this particular scenario, the model generated MSE values that came in at 0.00016, MAE values that came in at 0.00746, RMSE values that came in at 13.45, MAPE values that came in at 0.42, and R2 values that came in at 0.95. This is an indication that the model is functioning effectively. The conventional modeling techniques for forecasting, do not give the level of performance that is required. On the other hand, the results of this study will be useful for any type of time-specific forecasting prediction that requires a high level of accuracy.
- Subjects
MACHINE learning; ARTIFICIAL neural networks; DEEP learning; ARTIFICIAL intelligence; DECISION making
- Publication
Human-Centric Intelligent Systems, 2023, Vol 3, Issue 3, p275
- ISSN
2667-1336
- Publication type
Article
- DOI
10.1007/s44230-023-00039-x