TY - CONF TI - ClassBeacons: Designing Distributed Visualization of Teachers' Physical Proximity in the Classroom AU - An, Pengcheng AU - Bakker, Saskia AU - Ordanovski, Sara AU - Taconis, Ruurd AU - Eggen, Berry T3 - TEI '18 AB - As necessary for creating a learner-centered environment, nowadays teachers are expected to be more mindful about their proximity distribution: how to spend time in different locations of the classroom with individual learners. However feedback on this is only given to teachers by experts after classroom observation. In this paper we present the design and evaluation of ClassBeacons, a novel ambient information system that visualizes teachers' physical proximity through tangible devices distributed over the classroom. An expert review and a field evaluation with eight secondary school teachers were conducted to explore potential values of such a system and gather user experiences. Results revealed rich insights into how the system could influence teaching and learning, as well as how a distributed display can be seamlessly integrated into teachers' routines. C1 - New York, NY, USA C3 - Proceedings of the Twelfth International Conference on Tangible, Embedded, and Embodied Interaction DA - 2018/03/18/ PY - 2018 DO - 10.1145/3173225.3173243 DP - ACM Digital Library SP - 357 EP - 367 PB - Association for Computing Machinery SN - 978-1-4503-5568-1 ST - ClassBeacons UR - https://doi.org/10.1145/3173225.3173243 Y2 - 2021/03/07/00:00:00 KW - ambient information system KW - classroom KW - distributed display KW - learner-centered education KW - teacher proximity ER - TY - CONF TI - Automatic Teacher Modeling from Live Classroom Audio AU - Donnelly, Patrick J. AU - Blanchard, Nathan AU - Samei, Borhan AU - Olney, Andrew M. AU - Sun, Xiaoyi AU - Ward, Brooke AU - Kelly, Sean AU - Nystran, Martin AU - D'Mello, Sidney K. T3 - UMAP '16 AB - We investigate automatic analysis of teachers' instructional strategies from audio recordings collected in live classrooms. We collected a data set of teacher audio and human-coded instructional activities (e.g., lecture, question and answer, group work) in 76 middle school literature, language arts, and civics classes from eleven teachers across six schools. We automatically segment teacher audio to analyze speech vs. rest patterns, generate automatic transcripts of the teachers' speech to extract natural language features, and compute low-level acoustic features. We train supervised machine learning models to identify occurrences of five key instructional segments (Question & Answer, Procedures and Directions, Supervised Seatwork, Small Group Work, and Lecture) that collectively comprise 76% of the data. Models are validated independently of teacher in order to increase generalizability to new teachers from the same sample. We were able to identify the five instructional segments above chance levels with F1 scores ranging from 0.64 to 0.78. We discuss key findings in the context of teacher modeling for formative assessment and professional development. C1 - New York, NY, USA C3 - Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization DA - 2016/07/13/ PY - 2016 DO - 10.1145/2930238.2930250 DP - ACM Digital Library SP - 45 EP - 53 PB - Association for Computing Machinery SN - 978-1-4503-4368-8 UR - https://doi.org/10.1145/2930238.2930250 Y2 - 2021/03/07/00:00:00 KW - _C:Canada CAN KW - _C:Congo, Republic COG KW - _C:United States USA KW - __C:filed:1 KW - __C:scheme:1 KW - automatic feedback KW - classroom discourse KW - dialogic instruction KW - educational data mining KW - speech recognition ER -