Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment
Abstract Factor analysis (FA) receptor models are widely used for source apportionment (SA) due to their ability to extract the source contribution and profile from the data. However, there is subjectivity in the source identification and labelling due to manual interpretation, which is time-consumi...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2023-04-01
|
| Series: | Aerosol and Air Quality Research |
| Subjects: | |
| Online Access: | https://doi.org/10.4209/aaqr.220386 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823862856565653504 |
|---|---|
| author | Vikas Kumar Vasudev Malyan Manoranjan Sahu Basudev Biswal |
| author_facet | Vikas Kumar Vasudev Malyan Manoranjan Sahu Basudev Biswal |
| author_sort | Vikas Kumar |
| collection | DOAJ |
| description | Abstract Factor analysis (FA) receptor models are widely used for source apportionment (SA) due to their ability to extract the source contribution and profile from the data. However, there is subjectivity in the source identification and labelling due to manual interpretation, which is time-consuming. This raises a barrier to the development of the real-time SA process. In this study, a machine learning (ML) classification algorithm, k-nearest neighbour (kNN), is applied to the source profiles obtained from the United States Environmental Protection Agency’s (U.S. EPA) SPECIATE database to develop a model that can automatically label the factors derived from FA receptor models. The train and test score of the model is 0.85 and 0.79, respectively. The overall weighted average precision, recall and F1 score is 0.79. The performance of the model during validation exhibits acceptable results. The application of ML models for source profile labelling will reduce the time taken and the subjectivity associated with results due to modeler bias. This process can act as another layer of the process for verification of the results of FA receptor models. The application of this methodology advances the process towards real-time SA. |
| format | Article |
| id | doaj-art-f12c5a434c0f48529c345ab8e715a9d5 |
| institution | Kabale University |
| issn | 1680-8584 2071-1409 |
| language | English |
| publishDate | 2023-04-01 |
| publisher | Springer |
| record_format | Article |
| series | Aerosol and Air Quality Research |
| spelling | doaj-art-f12c5a434c0f48529c345ab8e715a9d52025-02-09T12:22:18ZengSpringerAerosol and Air Quality Research1680-85842071-14092023-04-0123711110.4209/aaqr.220386Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source ApportionmentVikas Kumar0Vasudev Malyan1Manoranjan Sahu2Basudev Biswal3Interdisciplinary Program in Climate Studies, Indian Institute of Technology BombayAerosol and Nanoparticle Technology Laboratory, Environmental Science and Engineering Department, Indian Institute of Technology BombayAerosol and Nanoparticle Technology Laboratory, Environmental Science and Engineering Department, Indian Institute of Technology BombayDepartment of Civil Engineering, Indian Institute of Technology BombayAbstract Factor analysis (FA) receptor models are widely used for source apportionment (SA) due to their ability to extract the source contribution and profile from the data. However, there is subjectivity in the source identification and labelling due to manual interpretation, which is time-consuming. This raises a barrier to the development of the real-time SA process. In this study, a machine learning (ML) classification algorithm, k-nearest neighbour (kNN), is applied to the source profiles obtained from the United States Environmental Protection Agency’s (U.S. EPA) SPECIATE database to develop a model that can automatically label the factors derived from FA receptor models. The train and test score of the model is 0.85 and 0.79, respectively. The overall weighted average precision, recall and F1 score is 0.79. The performance of the model during validation exhibits acceptable results. The application of ML models for source profile labelling will reduce the time taken and the subjectivity associated with results due to modeler bias. This process can act as another layer of the process for verification of the results of FA receptor models. The application of this methodology advances the process towards real-time SA.https://doi.org/10.4209/aaqr.220386Particulate matterSource apportionmentReceptor modelsMachine learningClassification |
| spellingShingle | Vikas Kumar Vasudev Malyan Manoranjan Sahu Basudev Biswal Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment Aerosol and Air Quality Research Particulate matter Source apportionment Receptor models Machine learning Classification |
| title | Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment |
| title_full | Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment |
| title_fullStr | Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment |
| title_full_unstemmed | Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment |
| title_short | Machine Learning Classification Model to Label Sources Derived from Factor Analysis Receptor Models for Source Apportionment |
| title_sort | machine learning classification model to label sources derived from factor analysis receptor models for source apportionment |
| topic | Particulate matter Source apportionment Receptor models Machine learning Classification |
| url | https://doi.org/10.4209/aaqr.220386 |
| work_keys_str_mv | AT vikaskumar machinelearningclassificationmodeltolabelsourcesderivedfromfactoranalysisreceptormodelsforsourceapportionment AT vasudevmalyan machinelearningclassificationmodeltolabelsourcesderivedfromfactoranalysisreceptormodelsforsourceapportionment AT manoranjansahu machinelearningclassificationmodeltolabelsourcesderivedfromfactoranalysisreceptormodelsforsourceapportionment AT basudevbiswal machinelearningclassificationmodeltolabelsourcesderivedfromfactoranalysisreceptormodelsforsourceapportionment |