Abstract:

Digital Storytelling can be an effective educational tool for teachers, but its most significant potential is for students who create their own stories. This type of activity can generate interest, attention, and motivation for the "digital generation" students in today’s classrooms The process can take advantage of the multimodal, creative talents of students as they begin to research and tell stories of their own while analyzing and synthesizing a wide range of content. One of the greatest advantages of using digital storytelling in the classroom is that it is a medium that is accessible to all types of learners. It does not require that students must communicate in writing, or visually, or that they are a confident public speaker. Storytelling engages all learning styles at a variety of levels. In addition, research shows that students learn most effectively when their emotions, as well as their intellect, are engaged.

This paper presents an approach with both digital storytelling and supervised machine learning of students’ engagement in STEM. Digital story telling are used to identify the factors that likely affect the students’ engagement. The supervised machine learning is modelled to predict the positive attitudes toward science, interests in STEM disciplines, and strengthen their intentions to pursue education pathways to science-related careers. The performance of model is evaluated through the decision tree and Naïve Bayes. Overall, the model has achieved a good prediction accuracy of 80.8%, along with limitations.

Keywords:

Engagement and Retention, Machine learning, Digital Storytelling.