Publication Date
2026
Presentation Length
Poster/Gallery presentation
College
College of Sciences & Mathematics
Department
Math and Computer Science, Department of
Student Level
Undergraduate
Faculty Mentor
Christina Davis
Presentation Type
Article
Summary
The beauty and skincare market offers thousands of products that often promise similar benefits while varying widely in price. Many consumers search for “dupes,” or lower-cost alternatives to high-end products, but finding them typically requires time-consuming research or unreliable online recommendations.
This project develops an AI/ML-based recommendation system that identifies similar beauty products using their ingredient compositions. Using a dataset of over 8,000 Sephora products, the system analyzes ingredient lists and product features to measure similarity and suggest potential alternatives. By combining similarity scores with pricing information, the model can highlight products that deliver comparable formulations at a lower cost.
The goal of this project is to explore whether ingredient-level data can effectively identify functionally similar products and reveal meaningful price differences across brands. This work demonstrates how artificial intelligence can be applied to improve consumer decision-making and make product discovery more efficient and accessible.
Recommended Citation
Tran, Jessica, "Finding Affordable Alternatives: A Similarity-Based Beauty Product Duplication" (2026). SPARK Symposium Presentations. 1102.
https://repository.belmont.edu/spark_presentations/1102
