- Titre : mobts : Mobility Time-Series Data Cleaning, Preprocessing, and Imputation
- Présentateur : Ali Shateri Benam (IFPEN)
- Résumé : There has been a surge in the analysis and utilization of mobility data, driven by the increasing availability of digitally recorded information. However, not all data are recorded with the same level of quality. Inevitably, issues arise from physical equipment responsible for data collection, compromising the accuracy, continuity, and consistency of recorded information. We put forward this work, presented in the reproducible form of a python package, to tackle this issue. The package provides a structured workflow for detecting measurement errors, handling missing observations, and imputing incomplete mobility time series. Beyond its technical contribution, this work aims to unify different cleaning and imputation approaches among collaborating researchers, providing an operational and reproducible tool for dealing with incomplete mobility data.
- Sujet de recherche associé : Action 133 - Investigate dynamic and heterogeneous behaviors of users with respect to their cyclic daily mobility