Facilitating the Open Learning and Education through Facial Analytics and Video Streaming

With the popularity of many massive online courses (MOOCs) and relevant platforms such as the Coursera, edX and Udemy, open learning and education will obviously become a very influential and significant sector in education all over the world, especially for developing countries like China and India with large population sizes and huge demands for tertiary education to develop the knowledge-based economy. Due to the potentially large class sizes for courses on an open learning platform, it can be difficult to monitor each individual or averaged performance in each open learning course, thus possibly leading to a relatively higher dropout rate at the end of the course. This is where computationally intelligent methods such as the machine learning or learning analytics techniques may help to effectively monitor each individual or group performance in an open learning course. In this work, we propose a cloud-based and personalized learning platform enhanced by a very efficient and intelligent facial analytics algorithm to capture learners' real-time responses when viewing any open educational resources such as the Ted Talks, YouTube or MERLOT on mobile devices. After analyzing learners' instant responses and attention spans, it may help to quickly identify those difficult and/or less interesting topics. Besides, interactive quizzes can be flexibly added by course instructors into different sections of an online educational resource to evaluate each individual learner's actual level of understanding while relevant video files or course material is being streamed onto the learner's mobile device from the Dropbox cloud storage. All the data about the learning progress will be securely uploaded onto a password-protected cloud computing platform for further analyzes. Furthermore, the cloud platform ensures the interoperability of various mobile devices in this newly enhanced learning analytic system with which some initial and positive students' feedback was collected. This work clearly provides many promising directions for future extensions of the next-generation open learning and education platform.