Proposing an AI-Based Approach to Raise Environmental Awareness

Abstract

Human activities, including industrial emissions, burning fossil fuels, and car emissions, are significant contributors to air pollution. In addition to harming the environment, it also poses an adverse impact on human health and the economy by releasing fine particles, nitrogen oxides, and sulfur dioxide into the atmosphere. According to the WHO, air pollution results in seven million preventable deaths yearly. By producing greenhouse gases like carbon dioxide and methane, air pollution also contributes to global warming. This causes biodiversity loss, rising sea levels, harsh weather, and food and water shortages. This research aims to inform the reader of the seriousness of environmental challenges via two stages since they have grown more severe over time. The LGBM algorithm produced a strong forecasting performance in the initial stage of predicting future air pollution. In the second stage, various machine learning algorithms were applied to the processed dataset, with Gaussian NB showing the highest accuracy among the six algorithms. The study comes to the conclusion that Gaussian NB is appropriate for high-dimensional data like the provided dataset and that machine learning algorithms may efficiently forecast air pollution, assisting in the development of associated policies. The pipes and suggested methodology could be helpful in addressing other comparable problems, including anticipating different pollutant levels or financial reporting. The paper’s overall argument is that the results will assist people in grasping the seriousness of environmental issues and that the anticipated outcomes may be demonstrated directly or through other environmental reports generated from the results.