Technologies for automated analysis of co-located, real-life, physical learning spaces

Abstract

The motivation for this paper is derived from the fact that there has been increasing interest among researchers and practitioners in developing technologies that capture, model and analyze learning and teaching experiences that take place beyond computer-based learning environments. In this paper, we review case studies of tools and technologies developed to collect and analyze data in educational settings, quantify learning and teaching processes and support assessment of learning and teaching in an automated fashion. We focus on pipelines that leverage information and data harnessed from physical spaces and/or integrates collected data across physical and digital spaces. Our review reveals a promising field of physical classroom analysis. We describe some trends and suggest potential future directions. Specifically, more research should be geared towards a) deployable and sustainable data collection set-ups in physical learning environments, b) teacher assessment, c) developing feedback and visualization systems and d) promoting inclusivity and generalizability of models across populations. CCS CONCEPTS• Human-centred computing~Visualization design and evaluation methods •Information systems~Data analytics KEYWORDSFace-to-face classroom analysis, co-located learning, physical learning analytics, educational data mining, educational technologies We utilized a three-pronged approach to select studies for review. First, relevant journals and conference proceedings, such as journals and proceedings of known conferences targeted at showcasing advancements and innovation in the field of learning sciences and learning analytics, were shortlisted and searched using a targeted search strategy. We include the following journals:

A Deep Transfer Learning Approach to Modeling Teacher Discourse in the Classroom

“…Teacher feedback systems generally serve one of two main purposes: (1) providing information on student learning (e.g., identifying at-risk students, real-time class orchestration); and (2) providing information on teaching pedagogy and effectiveness (e.g., improving professional development) [17,52]. Most systems are deployed in hybrid or virtual learning environments [17] and take advantage of extensive log data in the form of interactions with course materials, social interactions, assessment results, and time spent engaged with the platform [77]. For teachers in particular, the most common data recorded generally measures student time on platform and engagement with discussion boards [65].…”

Moodoo: Indoor Positioning Analytics for Characterising Classroom Teaching

“…There has been a growing interest in exploring physical aspects of the classroom [16]. For example, authors have used automated video analysis to model students' posture [45] and gestures [1], teacher's walking [10], interactions between teachers and students [1,53] during a lecture, and characterising the types of social interactions of students in makerspaces [15].…”

“…Each lab typically has between 30 and 40 students working in 10-13 small teams of 2-3 students each. Eighteen labs were randomly chosen (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18) for the study. All labs were conducted in the same (16.8 × 10 m) classroom equipped with workbenches, a lectern, a whiteboard, and multiple laboratory tools.…”

“…Classroom activity is ephemeral, and has largely remained opaque to computational analysis [38], with only a small number of artificial intelligence (AI) innovations targeting physical aspects of teaching and learning [16,38,47,54]. However, despite the online learning revolution, physical classrooms remain pervasive across all educational levels [7].…”

Where is the teacher? Digital analytics for classroom proxemics

“…There has been a growing interest in exploring physical aspects of the classroom using automated techniques (Chua, Dauwels, & Tan, 2019). For example, video analysis has been used as a way to identify students' posture and affective states during a lecture (Raca, Kidzinski, & Dillenbourg, 2015) or to quantify the number of interactions between lecturers and students (Watanabe, Ozeki, & Kohama, 2018).…”