Application of Machine Learning and Deep Learning in the analysis of air quality through IoT: a systematic review

Abstract

The growing concern about climate change and environmental deterioration has made air pollution a global problem. This article presents a systematic review on the application of Machine Learning and Deep Learning techniques in air quality assessment. The objective of this research is to evaluate the effectiveness of Machine Learning and Deep Learning techniques in air quality analysis, identifying those that show superior performance in terms of accuracy and reliability of results. The PRISMA method was applied to compile 65 relevant articles on air quality. The findings indicate that Machine Learning and Deep Learning are crucial in this area, especially in research from India and China. The most common methods in Machine Learning are SVM and Random Forest, while in Deep Learning LSTM and CNN stand out. It is concluded that Machine Learning and Deep Learning are essential to assess air quality using IoT, Machine Learning stands out for its accessibility and ease of interpretation in small data sets, while Deep Learning, despite requiring more resources and data, provides greater accuracy in the analysis.