In this paper, we discuss and address two common types of smartphone-based label errors:mislabeling and multi-action labels. We also compare multiple learning ...
Figures · 1) Inaccurate time stamps: A common problem with. labels is that the initial and end parts of a labeled activity. are incorrect. · 2) Mislabeling: An ...
label error in crowd-sourced smartphone-based sensor data and propose a machine learning model to detect error labels. We first collect 650 good labels and ...
In this paper, we discuss and address two common types of smartphone-based label errors:mislabeling and multi-action labels. We also compare multiple learning ...
Bibliographic details on Detecting Label Errors in Crowd-Sourced Smartphone Sensor Data.
ABSTRACT. Crowdsourcing is a challenging activity for many reasons, from task design to workers' training, identification of low-.
Missing: Smartphone | Show results with:Smartphone
117-121. Full Text via DOI: 10.1145/3197231.3197256. 2018. Detecting label errors in crowd-sourced smartphone sensor data. 20-25. Full Text via DOI: 10.1109 ...
Detecting label errors in crowd-sourced smartphone sensor data. X Bo, C Poellabauer, MK O'Brien, CK Mummidisetty, A Jayaraman. 2018 International Workshop on ...
Detecting Label Errors in Crowd-Sourced Smartphone Sensor Data. SocialSens ... Quantifying community mobility after stroke using mobile phone technology.