Search for: All records

Classroom sensing systems can capture data on teacher-student behaviours and interactions at a scale far greater than human observers can. These data, translated to multi-modal analytics, can provide meaningful insights to educational stakeholders. However, complex data can be difficult to make sense of. In addition, analyses done on these data are often limited by the organization of the underlying sensing system, and translating sensing data into meaningful insights often requires custom analyses across different modalities. We present Edulyze, an analytics engine that processes complex, multi-modal sensing data and translates them into a unified schema that is agnostic to the underlying sensing system or classroom configuration. We evaluate Edulyze’s performance by integrating three sensing systems (Edusense, ClassGaze, and Moodoo) and then present data analyses of five case studies of relevant pedagogical research questions across these sensing systems. We demonstrate how Edulyze’s flexibility and customizability allow us to answer a broad range of research questions made possible by Edulyze’s translation of a breadth of raw sensing data from different sensing systems into relevant classroom analytics.  

Ambient classroom sensing systems offer a scalable and non-intrusive way to find connections between instructor actions and student behaviors, creating data that can improve teaching and learning. While these systems effectively provide aggregate data, getting reliable individual student-level information is difficult due to occlusion or movements. Individual data can help in understanding equitable student participation, but it requires identifiable data or individual instrumentation. We propose ClassID, a data attribution method for within a class session and across multiple sessions of a course without these constraints. For within-session, our approach assigns unique identifiers to 98% of students with 95% accuracy. It significantly reduces multiple ID assignments compared to the baseline approach (3 vs. 167) based on our testing on data from 15 classroom sessions. For across-session attributions, our approach, combined with student attendance, shows higher precision than the state-of-the-art approach (85% vs. 44%) on three courses. Finally, we present a set of four use cases to demonstrate how individual behavior attribution can enable a rich set of learning analytics, which is not possible with aggregate data alone. 

Translating fine-grained activity detection (e.g., phone ring, talking interspersed with silence and walking) into semantically meaningful and richer contextual information (e.g., on a phone call for 20 minutes while exercising) is essential towards enabling a range of healthcare and human-computer interaction applications. Prior work has proposed building ontologies or temporal analysis of activity patterns with limited success in capturing complex real-world context patterns. We present TAO, a hybrid system that leverages OWL-based ontologies and temporal clustering approaches to detect high-level contexts from human activities. TAO can characterize sequential activities that happen one after the other and activities that are interleaved or occur in parallel to detect a richer set of contexts more accurately than prior work. We evaluate TAO on real-world activity datasets (Casas and Extrasensory) and show that our system achieves, on average, 87% and 80% accuracy for context detection, respectively. We deploy and evaluate TAO in a real-world setting with eight participants using our system for three hours each, demonstrating TAO's ability to capture semantically meaningful contexts in the real world. Finally, to showcase the usefulness of contexts, we prototype wellness applications that assess productivity and stress and show that the wellness metrics calculated using contexts provided by TAO are much closer to the ground truth (on average within 1.1%), as compared to the baseline approach (on average within 30%). 

The use of audio and video modalities for Human Activity Recognition (HAR) is common, given the richness of the data and the availability of pre-trained ML models using a large corpus of labeled training data. However, audio and video sensors also lead to significant consumer privacy concerns. Researchers have thus explored alternate modalities that are less privacy-invasive such as mmWave doppler radars, IMUs, motion sensors. However, the key limitation of these approaches is that most of them do not readily generalize across environments and require significant in-situ training data. Recent work has proposed cross-modality transfer learning approaches to alleviate the lack of trained labeled data with some success. In this paper, we generalize this concept to create a novel system called VAX (Video/Audio to 'X'), where training labels acquired from existing Video/Audio ML models are used to train ML models for a wide range of 'X' privacy-sensitive sensors. Notably, in VAX, once the ML models for the privacy-sensitive sensors are trained, with little to no user involvement, the Audio/Video sensors can be removed altogether to protect the user's privacy better. We built and deployed VAX in ten participants' homes while they performed 17 common activities of daily living. Our evaluation results show that after training, VAX can use its onboard camera and microphone to detect approximately 15 out of 17 activities with an average accuracy of 90%. For these activities that can be detected using a camera and a microphone, VAX trains a per-home model for the privacy-preserving sensors. These models (average accuracy = 84%) require no in-situ user input. In addition, when VAX is augmented with just one labeled instance for the activities not detected by the VAX A/V pipeline (~2 out of 17), it can detect all 17 activities with an average accuracy of 84%. Our results show that VAX is significantly better than a baseline supervised-learning approach of using one labeled instance per activity in each home (average accuracy of 79%) since VAX reduces the user burden of providing activity labels by 8x (~2 labels vs. 17 labels).