Search for: All records

AI recommendations shape our daily decisions and our young generation is no exception. The convenience of navigating personalized content comes with the notorious ‘‘filter bubble’’ effect, which can reduce exposure to diverse options and opinions. Children are particularly vulnerable to this due to their limited AI literacy and critical thinking skills. In this study, we explore how to engage children as co-designers to create child-centered experiences for learning AI concepts related to the filter bubble. Leveraging embodied and analogical learning theories, we co-designed an Augmented Reality (AR) application, BeeTrap, with children from underrepresented backgrounds in STEM. BeeTrap not only raises awareness of filter bubbles but also empowers children to understand recommendation system mechanisms. Our contributions include (1) insights into child-centered AI learning using embodied metaphors and analogies as educational representations of AI concepts; and (2) implications for enhancing children’s understanding of AI concepts through co-design processes. 

Understanding howAI recommendationswork can help the younger generation become more informed and critical consumers of the vast amount of information they encounter daily. However, young learners with limited math and computing knowledge often find AI concepts too abstract. To address this, we developed Briteller, a light-based recommendation system that makes learning tangible. By exploring and manipulating light beams, Briteller enables children to understand an AI recommender system’s core algorithmic building block, the dot product, through hands-on interactions. Initial evaluations with ten middle school students demonstrated the effectiveness of this approach, using embodied metaphors, such as "merging light" to represent addition. To overcome the limitations of the physical optical setup, we further explored how AR could embody multiplication, expand data vectors with more attributes, and enhance contextual understanding. Our findings provide valuable insights for designing embodied and tangible learning experiences that make AI concepts more accessible to young learners. 

Scripting is a widely-used way to automate the execution of tasks. Despite the popularity of scripting, it remains difficult to use for both beginners and experts: because of the cryptic commands for the first group, and incompatible syntaxes across different systems, for the latter group. The authors introduce Natural Shell, an assistant for enabling end-users to generate commands and scripts for various purposes. Natural Shell automatically synthesizes scripts for different shell systems based on natural language descriptions. By interacting with Natural Shell, new users can learn the basics of scripting languages without the obstacles from the incomprehensible syntaxes. On the other hand, the authors’ tool frees more advanced users from manuals when they switch shell systems. The authors have developed a prototype system and demonstrate its effectiveness with a benchmark of 50 examples of popular shell commands collected from online forums. In addition, the authors analyzed the usage of Natural Shell in a lab study that involves 10 participants with different scripting skill levels. Natural Shell effectively assists the users to generate commands in assigned syntaxes and greatly streamlines their learning and using experience.