BackHome
Refresh: OffView Personal Schedule
Session Summary

Automated Classroom Observation With Machine Perception

Fri, April 9, 10:40am to 12:10pm EDT (10:40am to 12:10pm EDT), SIG Sessions, SIG-Technology, Instruction, Cognition & Learning Paper and Symposium Sessions

Session Type: Symposium

Abstract

Multimodal data collection and analysis of school classrooms offer new opportunities to measure and improve instructional quality, but key challenges remain to employing them at scale. Recent technological advances in machine perception (i.e., machine learning and its applications to computer vision, speech recognition, natural language understanding, etc.) may provide ways of tackling these challenges. This symposium focuses on efforts to understand the role of machine perception in recognizing objects and activities in classroom audiovisual data and making judgments about instruction and teacher-student interactions. The included papers look across early childhood, elementary, and secondary classrooms, and across mathematics and English/language arts curricula, to explore ways that machine perception can make the process of analyzing audiovisual instructional data more accurate and efficient.

Sub Unit

Chair

Papers

Discussant