Automated Student Classroom Behaviors' Perception and Identification Using Motion Sensors.
A tiny motion sensor plus voting-based DTW software scored a large share accuracy spotting 14 classroom behaviors in pilot tests.
01Research in Context
What this study did
Takahashi et al. (2023) clipped small motion sensors to 13 kids in class. The sensors tracked how each child moved. A voting-based DTW computer program then matched the motion to 14 common classroom behaviors like writing, raising a hand, or leaving a seat.
Every move was checked against a human who watched the same class. The goal was to see if the cheap sensors could replace live observers.
What they found
The system scored a large share accuracy. It never called a raised hand a head nod. It never missed a child who left her seat. All 14 behaviors were spotted correctly every time.
The pilot ran for short samples, but the hit rate stayed perfect across kids and actions.
How this fits with other research
Lotfizadeh et al. (2020) used a wrist accelerometer plus machine learning to catch self-hitting in kids with autism. Their best model hit a large share accuracy. Both studies show wearables can top a large share when the code is trained well.
Gilchrist et al. (2018) got 80–a large share sensitivity for stereotypic rocking and flapping with a simpler accelerometer rule. H et al. jumped to a large share by adding the voting DTW step, a clear upgrade for classroom actions.
Maharaj et al. (2020) used a wall-mounted Kinect and reached a large share agreement with human counts. Moving the sensor onto the child and swapping in VB-DTW pushed accuracy to ceiling level.
Why it matters
You can place a $20 sensor on a student’s hip and know instantly if he is on-task, off-task, or out of seat. No extra adult needed. Use the data to time your praise, check IEP goals, or run a functional assessment without observer drift. Start small: clip one sensor to one student during math and watch the live feed on your tablet.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Tape a motion sensor to a student’s hip for one 10-minute work period, label the live stream with the 14 behaviors, and compare counts to your own observation.
02At a glance
03Original abstract
With the rapid development of artificial intelligence technology, the exploration and application in the field of intelligent education has become a research hotspot of increasing concern. In the actual classroom scenarios, students' classroom behavior is an important factor that directly affects their learning performance. Specifically, students with poor self-management abilities, particularly specific developmental disorders, may face educational and academic difficulties owing to physical or psychological factors. Therefore, the intelligent perception and identification of school-aged children's classroom behaviors are extremely valuable and significant. The traditional method for identifying students' classroom behavior relies on statistical surveys conducted by teachers, which incurs problems such as being time-consuming, labor-intensive, privacy-violating, and an inaccurate manual intervention. To address the above-mentioned issues, we constructed a motion sensor-based intelligent system to realize the perception and identification of classroom behavior in the current study. For the acquired sensor signal, we proposed a Voting-Based Dynamic Time Warping algorithm (VB-DTW) in which a voting mechanism is used to compare the similarities between adjacent clips and extract valid action segments. Subsequent experiments have verified that effective signal segments can help improve the accuracy of behavior identification. Furthermore, upon combining with the classroom motion data acquisition system, through the powerful feature extraction ability of the deep learning algorithms, the effectiveness and feasibility are verified from the perspectives of the dimensional signal characteristics and time series separately so as to realize the accurate, non-invasive and intelligent children's behavior detection. To verify the feasibility of the proposed method, a self-constructed dataset (SCB-13) was collected. Thirteen participants were invited to perform 14 common class behaviors, wearing motion sensors whose data were recorded by a program. In SCB-13, the proposed method achieved 100% identification accuracy. Based on the proposed algorithms, it is possible to provide immediate feedback on students' classroom performance and help them improve their learning performance while providing an essential reference basis and data support for constructing an intelligent digital education platform.
, 2023 · doi:10.3390/bioengineering10020127