Fine Grained Air Quality Monitoring Based on BiGRU and AM Model for Enhanced Public Health

Particles, biological molecules, and other harmful chemicals in the Earth's atmosphere generate air pollution, which in turn causes many negative health impacts, human deaths, animal deaths, agricultural failures, and infrastructure deterioration. Air pollution can be caused by both human actions and natural occurrences. Large-scale industry, which is relocating away from cities, is rarely held responsible for worsening air quality. Recent studies have found that automobile emissions account for the vast majority of urban air pollution. There are three steps to the proposed method: data preprocessing, feature selection, and finally training the model. Preprocessing in data science typically involves k-means and variational mode decomposition. Using the Q-learning-based swarm optimization method, features are chosen. When information gain is used for feature selection, BiGRU-AM is then used to train the model. When compared to GRU and BiGRU, two recognized alternatives, the proposed technique produces better overall results.