On Trusting Mobility Data Collected from Smartphones

Description

Solutions for smartphone-based travel surveys enabling automatic passive travel data collection are currently mushrooming, with most of them boasting accuracies of a least 90%. While the question of obtaining and organising sampling frames for trusted smart statistics from smartphone-based surveys is currently discussed in the transportation research community, there still exists a more fundamental problem concerning the rigorous assessment of the quality for the collected mobility data. This talk demonstrates the huge potential of passive mobility data collection with smartphones for reconstructing trips of people in terms of used travel modes and trip purposes, and points out fallacies and pitfalls regarding performance evaluation. Recent outcomes of an international activity recognition challenge show that results achieved `in-the-lab´ can easily become subject to an upward bias and cannot always serve as reliable indicators for the future performance `in-the-field´, where generality and robustness are essential.

Period	30 Jan 2019 → 31 Jan 2019
Event title	Eurostat Collaboration in Research and Methodology for Official Statistics: Workshop on Trusted Smart Statistics: policymaking in the age of the IoT
Event type	Other
Degree of Recognition	International