Source Apportionment of Particulate Matter by Application of Machine Learning Clustering Algorithms

A source apportionment (SA) study was conducted on two PM2.5 data sets, two carbon fractions and eight temperature-resolved carbon fractions collected during Cincinnati Childhood Allergy and Air Pollution Study (CCAAPS). This study aimed to evaluate two clustering algorithms: k-means clustering (kMC) and spectral clustering (SC) as potential receptor models for source apportionment. The application of kMC produced unsatisfactory results, but the results obtained from SC demonstrated a significant correlation with the results obtained using positive matrix factorization (PMF). The clustering results obtained were associated with practical evidence available in the literature. SC identified six source factors on analyzing two carbon fractions data set and seven factors from eight temperature-resolved carbon fractions data set. The sources (source contribution in parentheses) identified are: combustion (45.9 ± 3.66%) and secondary sulfate (11.4 ± 1.09%), vegetative/wood burning (17.5 ± 1.46%), diesel (10.6 ± 0.92%) and gasoline (3.6 ± 0.33%) vehicles, soil/crustal (2.07 ± 0.2%), traffic (9.3 ± 0.81%), and metal processing (8.8 ± 0.72%). The source profiles obtained using SC also show similarity with the profiles derived using PMF. In summary, this study presented a basic framework for applying Machine Learning algorithms for SA analysis. Also, it presents SC as a potential receptor model technique for SA.

• Publication year

  2022

• Venue

  Aerosol and Air Quality Research

• Publication date

  2022-01-27

• Fields of study

  Not labeled

• Identifiers
  DOI 10.4209/aaqr.210240
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Spectral Clustering

  2021cited by this paper
• Source apportionment of atmospheric particle number concentrations with wide size range by nonnegative matrix factorization (NMF).

  2021cited by this paper
• Global review of recent source apportionments for airborne particulate matter

  2020cited by this paper
• Efficient data preprocessing, episode classification, and source apportionment of particle number concentrations.

  2020cited by this paper
• Clustering Validation Measures

  2018cited by this paper
• Points of Significance: Clustering

  2017influential reference
• An Introduction to Machine Learning

  2017cited by this paper
• Review of receptor modeling methods for source apportionment

  2016influential reference
• Unsupervised Learning Algorithms

  2016cited by this paper
• Data Mining for Knowledge Discovery

  2015cited by this paper
• Contributions to cities' ambient particulate matter (PM): a systematic review of local source contributions at global level

  2015influential reference
• Trends of road dust emissions contributions on ambient air particulate levels at rural, urban and industrial sites in southern Spain

  2014influential reference
• An integrated PM2.5 source apportionment study: Positive Matrix Factorisation vs. the chemical transport model CAMx

  2014cited by this paper
• Assessment and application of clustering techniques to atmospheric particle number size distribution for the purpose of source apportionment

  2014cited by this paper
• Source apportionment of size segregated fine/ultrafine particle by PMF in Beijing

  2014cited by this paper
• The k-means clustering technique: General considerations and implementation in Mathematica

  2013cited by this paper
• Critical review and meta-analysis of ambient particulate matter source apportionment using receptor models in Europe

  2013influential reference
• Data Clustering : Algorithms and Applications

  2012influential reference
• Machine Learning - The Art and Science of Algorithms that Make Sense of Data

  2012cited by this paper
• Properties of aerosol signature size distributions in the urban environment as derived by cluster analysis

  2012cited by this paper
• Urban aerosol size distributions over the Mediterranean city of Barcelona, NE Spain

  2012cited by this paper
• Enhancing source apportionment with receptor models to foster the air quality directive implementation

  2012cited by this paper
• PM2.5 source apportionment: reconciling receptor models for U.S. nonurban and urban long-term networks.

  2011cited by this paper
• PMF analysis of wide-range particle size spectra collected on a major highway.

  2011cited by this paper
• Chemical compositions and source identification of PM₂.₅ aerosols for estimation of a diesel source surrogate.

  2011influential reference
• Semi-Supervised and Unsupervised Machine Learning: Novel Strategies

  2011cited by this paper
• Receptor model based identification of PM2.5 sources in Canadian cities

  2011influential reference
• Encyclopedia of Machine Learning

  2011influential reference
• Application of receptor modeling methods

  2011influential reference
• Principles and Theory for Data Mining and Machine Learning

  2010cited by this paper
• Sources of particulate matter: emission profile of biomass burning

  2009cited by this paper
• Quantifying road dust resuspension in urban environment by Multilinear Engine: A comparison with PMF2

  2009influential reference
• Cluster analysis of rural, urban, and curbside atmospheric particle size data.

  2009cited by this paper
• Principles and Theory for Data Mining and Machine Learning

  2009cited by this paper
• On evolutionary spectral clustering

  2009cited by this paper
• Handbook of Research on Machine Learning Applications and Trends : Algorithms , Methods , and Techniques

  2009influential reference
• Inter-comparison of receptor models for PM source apportionment: Case study in an industrial area

  2008influential reference
• A survey of kernel and spectral methods for clustering

  2008cited by this paper
• Fingerprinting particle origins according to their size distribution at a UK rural site

  2008cited by this paper
• Source apportionment of PM 2 . 5 : Comparing PMF and CMB results for four ambient monitoring sites in the southeastern United States

  2008cited by this paper
• Source apportionment of ambient fine particle size distribution using positive matrix factorization in Erfurt, Germany.

  2008cited by this paper
• A tutorial on spectral clustering

  2007cited by this paper
• Data clustering - theory, algorithms, and applications

  2007influential reference
• Source Apportionment of Air Particulate Matter by Chemical Mass Balance (CMB) and Comparison with Positive Matrix Factorization (PMF) Model

  2007cited by this paper
• Fundamental Limitations of Spectral Clustering

  2006cited by this paper
• Finding community structure in networks using the eigenvectors of matrices.

  2006cited by this paper
• UNMIX modeling of ambient PM(2.5) near an interstate highway in Cincinnati, OH, USA.

  2006cited by this paper
• A comparison between two receptor models to determine the source apportionment of atmospheric pollutants

  2006cited by this paper
• Selection of K in K-means clustering

  2005cited by this paper
• Clustering Methods

  2005cited by this paper
• Survey of clustering algorithms

  2005cited by this paper
• Spatial variability of fine particle mass, components, and source contributions during the regional air pollution study in St. Louis.

  2005influential reference
• Improving source identification of Atlanta aerosol using temperature resolved carbon fractions in positive matrix factorization

  2004influential reference
• The Elements of Statistical Learning: Data Mining, Inference, and Prediction

  2004cited by this paper
• Spectral methods for data analysis

  2004cited by this paper
• Application of receptor modeling to atmospheric constituents at Potsdam and Stockton, NY

  2003influential reference
• Learning Spectral Clustering

  2003cited by this paper
• Finding the Number of Clusters in a Dataset

  2003cited by this paper
• Comparison of Positive Matrix Factorization and Multilinear Engine for the source apportionment of particulate pollutants

  2003cited by this paper
• Source Apportionment of Phoenix PM2.5 Aerosol with the Unmix Receptor Model

  2003cited by this paper
• Review of PM2.5 and PM10 Apportionment for Fossil Fuel Combustion and Other Sources by the Chemical Mass Balance Receptor Model

  2002influential reference
• Receptor modeling application framework for particle source apportionment.

  2002cited by this paper
• On Clustering Validation Techniques

  2001cited by this paper
• On Spectral Clustering: Analysis and an algorithm

  2001cited by this paper
• A Random Walks View of Spectral Segmentation

  2001cited by this paper
• Sources of fine particle composition in the northeastern US

  2001influential reference
• Atmospheric aerosol over Vermont: chemical composition and sources.

  2001influential reference
• Identification of Sources of Phoenix Aerosol by Positive Matrix Factorization

  2000influential reference
• On clusterings-good, bad and spectral

  2000cited by this paper
• Data clustering: a review

  1999cited by this paper
• OPTICS: ordering points to identify the clustering structure

  1999cited by this paper
• Normalized cuts and image segmentation

  1997cited by this paper
• Elemental characterization of particulate matter emitted from biomass burning: Wind tunnel derived source profiles for herbaceous and wood fuels

  1997influential reference
• Least squares formulation of robust non-negative factor analysis

  1997cited by this paper
• From Data Mining to Knowledge Discovery in Databases

  1996cited by this paper
• BIRCH: an efficient data clustering method for very large databases

  1996cited by this paper
• A Second Course in Statistics: Regression Analysis

  1996cited by this paper
• In Advances in Neural Information Processing Systems

  1996cited by this paper
• An examination of procedures for determining the number of clusters in a data set

  1994cited by this paper
• Laplacian matrices of graphs: a survey

  1994cited by this paper
• Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values†

  1994cited by this paper
• The Handbook of Pattern Recognition and Computer Vision

  1993cited by this paper
• Handbook of Pattern Recognition and Computer Vision

  1993cited by this paper
• PARTITIONING SPARSE MATRICES WITH EIGENVECTORS OF GRAPHS*

  1990cited by this paper
• in Advances in Neural Information Processing

  1989cited by this paper
• Cluster analysis of electron microprobe analysis data of individual particles for source apportionment of air particulate matter

  1988cited by this paper
• Least squares quantization in PCM

  1982influential reference
• A Cluster Separation Measure

  1979cited by this paper
• A dendrite method for cluster analysis

  1974cited by this paper
• Some methods for classification and analysis of multivariate observations

  1967influential reference
• Cluster analysis of multivariate data : efficiency versus interpretability of classifications

  1965cited by this paper
• On the generalized distance in statistics

  1936cited by this paper

• Advances in air quality modeling through artificial intelligence, machine learning, and deep learning: A comprehensive review.

  2026cites this paper
• Temporal Dynamics and Source Characteristics of Fine Particulate Matter using Positive Matrix Factorization (PMF)

  2025cites this paper
• Machine learning analysis of drivers of differences in PAH content between PM1 and PM10 in Zagreb, Croatia

  2025cites this paper
• Tracing Petrochemical Emissions: Neural Network-Driven Spectral Fingerprint Analysis of Volatile Organic Compounds

  2025cites this paper
• Source apportionment and health risks of heavy metals in agricultural soils near mining areas: APCS-MLR and Monte Carlo approaches

  2025cites this paper
• Unsupervised machine learning to investigate the joint effects of SO2, NO2, O3, PM2.5 and PM10 on respiratory and cardiovascular hospital admissions in the Vaal Triangle Airshed Priority Area, South Africa.

  2025cites this paper
• Exploring AI Potential in Optical Analysis of Organic and Elemental Carbons: A Review of Emerging Applications in Airborne Particulate Characterization

  2025cites this paper
• Assessing the Spatial Transferability of Calibration Models across a Low-cost Sensors Network

  2024cites this paper
• Real-Time Source Apportionment of Particulate Matter from Low-Cost Particle Sensors Using Machine Learning

  2024cites this paper
• Aerosols in Northern Morocco (Part 3): the application of three complementary approaches towards a better understanding of PM10 sources

  2023cites this paper
• Application of machine learning in atmospheric pollution research: A state-of-art review.

  2023cites this paper
• A machine learning modelling approach to characterize the background pollution in the Western Macedonia region in northwest Greece

  2023cites this paper