MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES
Cities worldwide are attempting to be claimed as smart, but truly classifying as such remains a great challenge. This paper aims to use artificial intelligence AI to classify the smart city's performance as well as the factors linked to it. This is based on the perceptions of residents on issue...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2024-03-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/221/2024/isprs-archives-XLVIII-4-W9-2024-221-2024.pdf |
| _version_ | 1827322555464155136 |
|---|---|
| author | L. Hammoumi H. Rhinane |
| author_facet | L. Hammoumi H. Rhinane |
| author_sort | L. Hammoumi |
| collection | DOAJ |
| description | Cities worldwide are attempting to be claimed as smart, but truly classifying as such remains a great challenge. This paper aims to use artificial intelligence AI to classify the smart city's performance as well as the factors linked to it. This is based on the perceptions of residents on issues related to structures and technology applications available in their cities. To achieve this goal, the study included 200 cities worldwide. For 147 cities we captured the perceptions of 120 residents in each city, by answering a survey of 39 questions evolving around two main Pillars: ‘Structures’ that refers to the existing infrastructure of the city and the ‘Technology’ pillar that describes the technological provisions and services available to the inhabitants. And each one is evaluated under five key areas: health and safety, mobility, activities, opportunities, and governance. The final score of the other 53 cities, was measured by using the data openly available on the internet. And this by means of different algorithms of machine learning such as Random Forest RF, Artificial Neural Network ANN, Support Vector Machine (SVM), and Gradient Boost (XGB). These algorithms have been compared and evaluated in order to select the best one. The tests showed that Random Forest RF alongside with Artificial Neural Network ANN, with the highest level of accuracy, are the best trained model. This study will enable other researches to use machine learning in the identification process of smart cities. |
| first_indexed | 2024-04-25T01:27:09Z |
| format | Article |
| id | doaj.art-b745ecb95827416ebdaf6596ab89b79f |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-04-25T01:27:09Z |
| publishDate | 2024-03-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-b745ecb95827416ebdaf6596ab89b79f2024-03-08T19:22:10ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342024-03-01XLVIII-4-W9-202422122810.5194/isprs-archives-XLVIII-4-W9-2024-221-2024MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIESL. Hammoumi0H. Rhinane1Geoscience Laboratory, Aïn Chock Faculty of Science, University Hassan II, Casablanca, MoroccoGeoscience Laboratory, Aïn Chock Faculty of Science, University Hassan II, Casablanca, MoroccoCities worldwide are attempting to be claimed as smart, but truly classifying as such remains a great challenge. This paper aims to use artificial intelligence AI to classify the smart city's performance as well as the factors linked to it. This is based on the perceptions of residents on issues related to structures and technology applications available in their cities. To achieve this goal, the study included 200 cities worldwide. For 147 cities we captured the perceptions of 120 residents in each city, by answering a survey of 39 questions evolving around two main Pillars: ‘Structures’ that refers to the existing infrastructure of the city and the ‘Technology’ pillar that describes the technological provisions and services available to the inhabitants. And each one is evaluated under five key areas: health and safety, mobility, activities, opportunities, and governance. The final score of the other 53 cities, was measured by using the data openly available on the internet. And this by means of different algorithms of machine learning such as Random Forest RF, Artificial Neural Network ANN, Support Vector Machine (SVM), and Gradient Boost (XGB). These algorithms have been compared and evaluated in order to select the best one. The tests showed that Random Forest RF alongside with Artificial Neural Network ANN, with the highest level of accuracy, are the best trained model. This study will enable other researches to use machine learning in the identification process of smart cities.https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/221/2024/isprs-archives-XLVIII-4-W9-2024-221-2024.pdf |
| spellingShingle | L. Hammoumi H. Rhinane MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES |
| title_full | MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES |
| title_fullStr | MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES |
| title_full_unstemmed | MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES |
| title_short | MACHINE LEARNING (AI) FOR IDENTIFYING SMART CITIES |
| title_sort | machine learning ai for identifying smart cities |
| url | https://isprs-archives.copernicus.org/articles/XLVIII-4-W9-2024/221/2024/isprs-archives-XLVIII-4-W9-2024-221-2024.pdf |
| work_keys_str_mv | AT lhammoumi machinelearningaiforidentifyingsmartcities AT hrhinane machinelearningaiforidentifyingsmartcities |