PALS: Personalized Adaptive Learning System

With the advent of the Internet-age, transfer of information between individuals has become spontaneous, effective, and economical. The prevalence of social media in our daily lives has spurred knowledge sharing to enhance our experiences. In addition, the rise of video sharing platforms such as TikTok shows a trend to share knowledge through videos rather than written media. However, in in-person classes, the prevalent model still seems to be instructor-centered, where knowledge primarily flows from the instructor to students, and students are expected to gather missing information primarily on their own. Students in large classrooms, especially in public universities, experience a large bias in their learning experience; students closer to the instructors' knowledge gaining the most while the middle and the tail end of the student-body are left to fend for themselves. Instructors, on the other hand, cannot be expected to accommodate each individual in their class. They neither have the time nor the resources required to hand-hold each student through each question for a course. The primary goal of this paper is to propose Personalized Adaptive Learning System(PALS), a scalable video-based social learning network utilizing student-sourced; student-verified content aimed at guiding students through each question in a course to help each student chart a path from what they know to what they want to know by engaging them in various modes of learning: watching, solving, teaching, and evaluating video solutions. Consequently, this system bridges the knowledge transmission gap that exists in traditional classrooms. To deal with the large amount of content, a Charisma scale is proposed to filter good quality student-sourced videos.