Understanding people’s movements in urban areas is critical for efficient urban planning and infrastructure management in smart cities. In this work, we analyze the crowding dynamics in six urban areas of the municipality of Brescia characterized by different land use using recent high technology data. Recurring be- haviors in the days of a year are captured by means of functional data clustering. Two clustering methodologies are applied: the k-means Alignment, and the Model- Based Functional Data Clustering. The results of the two procedures are compared using the Rand index and by analyzing the clusters’ similarities over a set of passive descriptive variables.
Comprendere gli spostamenti delle persone nelle aree urbane `e fonda- mentale per un’efficiente pianificazione urbana e gestione delle infrastrutture nelle smart cities. In questo lavoro si analizzano le dinamiche di affollamento in sei aree urbane del comune di Brescia caratterizzate da diverso uso del suolo utilizzando dati recenti ad alta tecnologia. I comportamenti ricorrenti nei giorni di un anno vengono catturati mediante clustering di dati funzionali. Vengono applicate due metodologie di clustering: il k-means Alignment, ed il Model-Based Functional Data Clustering. I risultati delle due procedure vengono confrontati utilizzando l’indice di Rand e analizzando le somiglianze dei cluster su un insieme di variabili descrittive passive.
(2023). Functional clustering methods for space-time big data from mobile phone networks . Retrieved from https://hdl.handle.net/10446/249609
Functional clustering methods for space-time big data from mobile phone networks
Metulini, Rodolfo;Carpita, Maurizio
2023-01-01
Abstract
Understanding people’s movements in urban areas is critical for efficient urban planning and infrastructure management in smart cities. In this work, we analyze the crowding dynamics in six urban areas of the municipality of Brescia characterized by different land use using recent high technology data. Recurring be- haviors in the days of a year are captured by means of functional data clustering. Two clustering methodologies are applied: the k-means Alignment, and the Model- Based Functional Data Clustering. The results of the two procedures are compared using the Rand index and by analyzing the clusters’ similarities over a set of passive descriptive variables.File | Dimensione del file | Formato | |
---|---|---|---|
IES2023_shortversion_perazzini_et_al.pdf
Solo gestori di archivio
Versione:
publisher's version - versione editoriale
Licenza:
Licenza default Aisberg
Dimensione del file
652.19 kB
Formato
Adobe PDF
|
652.19 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo