Programming, Education, and Computer-Human Interaction Lab (Prof. April Wang)

Open Opportunities

Collaborative Data Storytelling in Spreadsheets with Large Language Models ETH Zurich Programming, Education, and Computer-Human Interaction Lab (Prof. April Wang) This project aims to transform the way users engage with data in spreadsheets by creating an interactive, collaborative platform for data storytelling. The motivation is to make data analysis more engaging, accessible, and narrative-driven, allowing users to seamlessly weave stories around their data while leveraging the power of large language models (LLMs). The platform will feature a community-based web forum where users can access a variety of public datasets or upload their own. Each dataset can be imported into a storytelling system integrated within the spreadsheet environment. This system will allow users to perform common data operations, such as calculating averages, standard deviations, and generating visualizations (e.g., charts), while the LLM will offer suggestions for a cohesive narrative. For instance, the system can auto-generate explanatory text based on the user’s data manipulations, creating a comprehensive story of the dataset. The goal is to help users articulate insights, uncover trends, and communicate findings in a more structured and compelling way. (Target Venue: ACM UIST 2025) Interdisciplinary Engineering ETH Zurich (ETHZ), Master Thesis

Understanding Youth & Social Media: User Generated Content and Algorithm Bias ETH Zurich Programming, Education, and Computer-Human Interaction Lab (Prof. April Wang) This project seeks to raise awareness among young people about the presence and influence of algorithmic bias in social media apps. Youth often interact with these platforms without fully understanding how their interactions contribute to biased content curation. To address this, we propose the development of a web-based system that simulates a social media app, providing real-time feedback on how user-generated content can influence algorithmic bias. The web system will function as a mock social media platform where users, particularly young people, can post content as they would on real social media apps. When users submit posts, the system will generate a preview that shows how the content spreads through the platform. This preview will visually represent how the content engages with algorithms, potentially amplifying biases based on factors like language, content type, or context. By seeing this in action, users will learn how their posts can unintentionally reinforce algorithmic bias, encouraging them to be more mindful of their contributions. (Target Venue: ACM UIST 2025) Interdisciplinary Engineering Bachelor Thesis, ETH Zurich (ETHZ), Master Thesis