Proceedings of the 2014 ACM workshop on Multimodal Learning Analytics Workshop and Grand Challenge

Abstract

Learning Analytics is the "middle-space" where Educational Sciences, Computer Science, Learning Technologies and Data Science converge. The main goal of this new field of knowledge is to contribute to new empirical findings, theories, methods, and metrics for understanding how students learn and to use that knowledge to improve those students' learning. Multimodal Learning Analytics, which emphasizes the analysis of natural rich modalities of communication during situated learning activities, is one of the most challenging but, at time, more promising areas of Learning Analytics. The Third International Workshop on Multimodal Learning Analytics brings together researchers in multimodal interaction and systems, cognitive and learning sciences, educational technologies, and related areas to discuss the recent developments and future opportunities in this sub-field. Following the First International Workshop on Multimodal Learning Analytics in Santa Monica in 2012 and the ICMI Grand Challenge on Multimodal Learning Analytics in Sydney in 2013, this third workshop comprises a mixture of a workshop session and two data-driven grand challenges. The program committee reviewed and accepted the following articles. The workshop session focuses on the presentation of multimodal signal analysis techniques that could be applied in Multimodal Learning Analytics. In this workshop challenges presenters concentrate on the benefits and shortcomings of different research and technical methods used for multimodal analysis of learning signals. This session includes four articles from diverse topics: theoretical and conceptual considerations for different forms of multimodal data fusion; voice analysis to determine the level of rapport in learning exercises; video analysis of live classrooms; and the role of multimodal analysis in the service of studying complex learning environments. Following the successful experience of the previous Multimodal Learning Analytics Grand Challenge in ICMI 2013, this year, this event will provide two data sets with a wealth of research questions to be tackled by interested participants: Math Data Challenge and Presentation Quality Challenge. For the Math Data Challenge, one article presented in this session provides a detailed exploration of how to use the digital pen information to predict the expertise in the group. This work reaches high levels of accuracy (83%) when identifying the expert student among the participants. For the Presentation Quality Challenge three articles are presented. The first one explores the slide presentation files and the audio features to predict the grade obtained by each student. The second work makes use of all the provided modalities (audio, video, Kinect data and slide files) and suggests that multimodal cues can predict human scores on presentation tasks. The final article uses the video and Kinect information to predict human grading. The third Multimodal Learning Analytics Workshop and Grand Challenges (MLA'14) was envisioned as a venue to initiate research in this nascent subfield of Learning Analytics. New challenges and insights will arise from the convergence of practitioners, academics and researchers, which in turn will create opportunities to collaborate and to create applications and tools to assist students, teachers and the community.