A supervised approach for land use identification in Trento using mobile phone data as an alternative to unsupervised clustering techniques

Abstract

This study explores land-use classification in Trento using supervised learning and call detail records as a proxy for human activity. The alpine setting introduces varied terrain and sparse network coverage. Comparative experiments with k-nearest neighbours, support vector machines and random forests show that supervised approaches can capture complex spatiotemporal patterns more accurately than unsupervised clustering. The results demonstrate the potential of mobile-network data for cost-effective, fine-grained land-use monitoring across urban, agricultural and forested areas.

Citation

Mendoza-Hurtado, M., Cerruela-García, G., & Ortiz-Boyer, D. (2025). A supervised approach for land use identification in Trento using mobile phone data as an alternative to unsupervised clustering techniques. Applied Sciences, 15(4), Article 1753. https://doi.org/10.3390/app15041753