%0 Journal Article %1 9965572 %A Echeverria, Vanessa %A Martinez-Maldonado, Roberto %A Yan, Lixiang %A Zhao, Linxuan %A Fernandez-Nieto, Gloria %A Gašević, Dragan %A Shum, Simon Buckingham %D 2022 %J IEEE Pervasive Computing %K analytics collaboration learning listening multimodal sensors smart space %P 1-11 %R 10.1109/MPRV.2022.3217454 %T HuCETA: A Framework for Human-Centered Embodied Teamwork Analytics %U https://ieeexplore.ieee.org/document/9965572 %X Collocated teamwork remains a pervasive practice across all professional sectors. Even though live observations and video analysis have been utilized for understanding embodied interaction of team members, these approaches are impractical for scaling up the provision of feedback that can promote developing high-performance teamwork skills. Enriching spaces with sensors capable of automatically capturing team activity data can improve learning and reflection. Yet, connecting the enormous amounts of data such sensors can generate with constructs related to teamwork remains challenging. This article presents a framework to support the development of human-centered embodied teamwork analytics by 1) enabling hybrid human–machine multimodal sensing; 2) embedding educators’ and experts’ knowledge into computational team models; and 3) generating human-driven data storytelling interfaces for reflection and decision making. This is illustrated through an in-the-wild study in the context of healthcare simulation, where predictive modeling, epistemic network analysis, and data storytelling are used to support educators and nursing teams.

@article{9965572, abstract = {Collocated teamwork remains a pervasive practice across all professional sectors. Even though live observations and video analysis have been utilized for understanding embodied interaction of team members, these approaches are impractical for scaling up the provision of feedback that can promote developing high-performance teamwork skills. Enriching spaces with sensors capable of automatically capturing team activity data can improve learning and reflection. Yet, connecting the enormous amounts of data such sensors can generate with constructs related to teamwork remains challenging. This article presents a framework to support the development of human-centered embodied teamwork analytics by 1) enabling hybrid human–machine multimodal sensing; 2) embedding educators’ and experts’ knowledge into computational team models; and 3) generating human-driven data storytelling interfaces for reflection and decision making. This is illustrated through an in-the-wild study in the context of healthcare simulation, where predictive modeling, epistemic network analysis, and data storytelling are used to support educators and nursing teams.}, added-at = {2022-12-26T09:18:13.000+0100}, author = {Echeverria, Vanessa and Martinez-Maldonado, Roberto and Yan, Lixiang and Zhao, Linxuan and Fernandez-Nieto, Gloria and Gašević, Dragan and Shum, Simon Buckingham}, biburl = {https://www.bibsonomy.org/bibtex/2a3e011148ab6ae35dcc7c65418e1f6f4/yish}, doi = {10.1109/MPRV.2022.3217454}, interhash = {c4e6310bf0b9451f0ee1b55b5c171d8a}, intrahash = {a3e011148ab6ae35dcc7c65418e1f6f4}, issn = {1558-2590}, journal = {IEEE Pervasive Computing}, keywords = {analytics collaboration learning listening multimodal sensors smart space}, pages = {1-11}, timestamp = {2022-12-26T09:18:13.000+0100}, title = {HuCETA: A Framework for Human-Centered Embodied Teamwork Analytics}, url = {https://ieeexplore.ieee.org/document/9965572}, year = 2022 }