Gift shopping can be challenging due to limited knowledge of the recipient's preferences, often leading to after-purchase regret. Our project explores the effectiveness of Fashion Recommender Systems (FRS) in the context of gift purchases with limited preference data. We conducted experiments to compare human and FRS recommendations for fashion gift buyers, revealing that both can achieve significant correctness in their predictions. Our research also highlights the potential of FRS in cold-start scenarios and investigates gender-based differences in recommendation accuracy. The findings suggest that incorporating FRS into real-world applications, like in-store shopping, can improve the gift-buying experience.
We designed a smartphone app to demonstrate the integration of FRS into practical applications, helping users choose fashion gifts with limited data on recipient preferences. This repository contains code, datasets, results from our experiments, and the app prototype.
This project has contributed to the following academic paper:
- The Art of Gift-Giving with Limited Preference Data: How Fashion Recommender Systems Can Help
- Authors: Sharareh Alipour, Sina Elahimanesh, Aliakbar Ghayouri Sales, Iman Mohammadi, Parimehr Morassafar, Seyed Parsa Neshaei
- Conference: CHI '24: CHI Conference on Human Factors in Computing Systems
- Date: May 2024
- DOI: 10.1145/3613905.3651000
