A Scalable Method for Rapid and Efficient Skill Discovery for Next-Generation Adaptive Online Courseware

• In Event: Paper Session: Innovations in Online Instruction

Abstract

Massive Open Online Courses (MOOCs) have expanded their influence in the education field, setting a new trend in technology and pedagogical areas. However, they have controversial issues such as high dropout rates and ineffective and unqualified course environments. Especially, it has been argued that MOOCs should provide an adaptive online learning environment to support diverse students’ efficient learning. The purpose of this paper is to propose a scalable method for hybrid skill modelling, an innovative way to design a competency-driven adaptive online course. Employing both human-labor skill mapping and machine-generated skill mapping with text-mining technology helped us extract a set of skills (a skill model) from which we drove instructions and assessments to provide synergetic learning.

Authors

• Seoyeon Park, Texas A&M University - College Station

• Abhishek Sharma, Texas A&M University

• Donna Miller, Bryan Independent School District

• Noboru Matsuda, Texas A&M University - College Station