Three novel cost-sensitive machine learning models for urban growth modelling
This article addresses the class imbalance problem in urban gain modelling (UGM) of Tabriz and Isfahan megacities in Iran by proposing novel cost-sensitive machine learning models, namely cost-sensitive support vector machine (CSVM), random forest (CRF) and artificial neural network (CANN). Random s...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-01-01
|
| Series: | Geocarto International |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10106049.2024.2353252 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850108131703521280 |
|---|---|
| author | Mohammad Ahmadlou Mohammad Karimi Saad Sh. Sammen Karam Alsafadi |
| author_facet | Mohammad Ahmadlou Mohammad Karimi Saad Sh. Sammen Karam Alsafadi |
| author_sort | Mohammad Ahmadlou |
| collection | DOAJ |
| description | This article addresses the class imbalance problem in urban gain modelling (UGM) of Tabriz and Isfahan megacities in Iran by proposing novel cost-sensitive machine learning models, namely cost-sensitive support vector machine (CSVM), random forest (CRF) and artificial neural network (CANN). Random sampling, a frequently utilized method, fails to effectively tackle this issue by biasing models towards no change samples, which outnumber change samples. The results showed that CRF exhibited the highest accuracy (AUC = 0.560), followed by CANN (AUC = 0.557) and CSVM (AUC = 0.448) in Isfahan. In Tabriz, CRF (AUC = 0.809) and CANN (AUC = 0.818) excelled, outperforming balanced sampling models constructed with ANN, RF and SVM with the AUROC of ANN and RF boosted by 15% and 2% in validation. By emphasizing the significance of addressing class imbalance appropriately, this research highlights the improvement in modelling outcomes achievable through the cost-sensitive models especially in Tabriz case. |
| format | Article |
| id | doaj-art-ead5aef79d2f4eb795cf779d4683a150 |
| institution | OA Journals |
| issn | 1010-6049 1752-0762 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geocarto International |
| spelling | doaj-art-ead5aef79d2f4eb795cf779d4683a1502025-08-20T02:38:26ZengTaylor & Francis GroupGeocarto International1010-60491752-07622024-01-0139110.1080/10106049.2024.2353252Three novel cost-sensitive machine learning models for urban growth modellingMohammad Ahmadlou0Mohammad Karimi1Saad Sh. Sammen2Karam Alsafadi3GIS Department, Geodesy and Geomatics Faculty, K.N. Toosi University of Technology, Tehran, IranGIS Department, Geodesy and Geomatics Faculty, K.N. Toosi University of Technology, Tehran, IranDepartment of Civil Engineering, College of Engineering, University of Diyala, Baqubah, IraqDepartment of Geography, Faculty of Arts and Humanities, Damascus University, Damascus, SyriaThis article addresses the class imbalance problem in urban gain modelling (UGM) of Tabriz and Isfahan megacities in Iran by proposing novel cost-sensitive machine learning models, namely cost-sensitive support vector machine (CSVM), random forest (CRF) and artificial neural network (CANN). Random sampling, a frequently utilized method, fails to effectively tackle this issue by biasing models towards no change samples, which outnumber change samples. The results showed that CRF exhibited the highest accuracy (AUC = 0.560), followed by CANN (AUC = 0.557) and CSVM (AUC = 0.448) in Isfahan. In Tabriz, CRF (AUC = 0.809) and CANN (AUC = 0.818) excelled, outperforming balanced sampling models constructed with ANN, RF and SVM with the AUROC of ANN and RF boosted by 15% and 2% in validation. By emphasizing the significance of addressing class imbalance appropriately, this research highlights the improvement in modelling outcomes achievable through the cost-sensitive models especially in Tabriz case.https://www.tandfonline.com/doi/10.1080/10106049.2024.2353252Imbalance problemcost-sensitive algorithmsurban gain modellingrandom samplingdata miningmachine learning model |
| spellingShingle | Mohammad Ahmadlou Mohammad Karimi Saad Sh. Sammen Karam Alsafadi Three novel cost-sensitive machine learning models for urban growth modelling Geocarto International Imbalance problem cost-sensitive algorithms urban gain modelling random sampling data mining machine learning model |
| title | Three novel cost-sensitive machine learning models for urban growth modelling |
| title_full | Three novel cost-sensitive machine learning models for urban growth modelling |
| title_fullStr | Three novel cost-sensitive machine learning models for urban growth modelling |
| title_full_unstemmed | Three novel cost-sensitive machine learning models for urban growth modelling |
| title_short | Three novel cost-sensitive machine learning models for urban growth modelling |
| title_sort | three novel cost sensitive machine learning models for urban growth modelling |
| topic | Imbalance problem cost-sensitive algorithms urban gain modelling random sampling data mining machine learning model |
| url | https://www.tandfonline.com/doi/10.1080/10106049.2024.2353252 |
| work_keys_str_mv | AT mohammadahmadlou threenovelcostsensitivemachinelearningmodelsforurbangrowthmodelling AT mohammadkarimi threenovelcostsensitivemachinelearningmodelsforurbangrowthmodelling AT saadshsammen threenovelcostsensitivemachinelearningmodelsforurbangrowthmodelling AT karamalsafadi threenovelcostsensitivemachinelearningmodelsforurbangrowthmodelling |