Tekouabou, Stephane Cedric KoumetioChenal, JeromeAzmi, RidaDiop, El BachirToulni, Hamza2023-06-052023-06-052023-06-052022-01-0110.1007/978-3-031-22950-3_16https://infoscience.epfl.ch/handle/20.500.14299/198004WOS:000982335900016Urban data strongly favored by urban digitization and web 2.0 constitute the raw material whose sources often diverge. However, as far as we know, little or no work has been done to explore the sources and targeted applications for urban planning indicators modelling. We aim to guide neophytes who seek to integrate smart data-driven applications into urban planning processes with greater clarity and credibility. For this purpose, we used test mining to analyze 250 (out of more than 750) relevant papers in the Scopus database and applied ML to an urban planning problem. We found that the data comes broadly from two main categories of sources, namely sensors and statistical surveys (including social network data). Data sources are highly correlated with their structure and the potential planning issues addressed. We conclude our work by discussing the potentialities, emerging issues, and challenges that urban data sources should face to better catalyze intelligent planning.Computer Science, Information SystemsComputer Science, Interdisciplinary ApplicationsComputer Scienceurban datadata sourceurban planningsmart citiesmachine learningurban planning indicatorsurban modellingbig datamachineregressionorangeformtext::conference output::conference proceedings::conference paper