Litcius/Paper detail

Applying machine learning techniques in air quality prediction

Екатерина Николаевна Гладкова, Liliya Saychenko

2022Transportation research procedia45 citationsDOIOpen Access PDF

Abstract

Air pollution levels become more and more dangerous every year, which is one оf the most significant and important problems for humanity nowadays. Of all air pollutants, particulate matter has the most harmful effect on the human health, especially in the long term. This article discusses the need to predict changes in PM2.5 concentrations for air quality monitoring and preventing unsafe and risky situations. In the course of the presented study, detailed visualization of the original data and а comparative survey of the application of machine learning methods were carried out. For solving the problem of forecasting time series of particulate matter concentration values several machine learning models, such as ARIMA, Facebook Prophet, and LSTM, were used. At the moment, it is possible to predict the trend of changes in the mean values of pollutant concentrations for several months in advance. But widespread use of these technologies requires more correct data to make better and more accurate predictions.

Topics & Concepts

Autoregressive integrated moving averageAir quality indexParticulatesMachine learningAir pollutionArtificial intelligenceAir pollutantsPollutantComputer scienceQuality (philosophy)Moment (physics)Big dataEnvironmental scienceData miningTime seriesMeteorologyGeographyChemistryEpistemologyClassical mechanicsOrganic chemistryPhysicsPhilosophyAir Quality Monitoring and ForecastingAdvanced Computational Techniques in Science and EngineeringStatistical and Computational Modeling
Applying machine learning techniques in air quality prediction | Litcius