| Title |
Jacquard Similarity-based Hairstyle Recommendation System based on Face Shapes |
| Authors |
이상민(Sangmin Lee) ; 정재형(Jaehyung Jung) ; 이계식(Gyesik Lee) |
| DOI |
https://doi.org/10.5573/ieie.2025.62.4.47 |
| Keywords |
Jacquard similarity; CNN; Hairstyle recommendation system; Data analysis |
| Abstract |
As the demand for personalized recommendation services in various industries continues to grow, the importance of sophisticated recommendation algorithms that incorporate user preferences and physical features has become increasingly important. In this study, we propose a novel hairstyle recommendation system based on Jacquard similarity to weight users’ facial features and style preferences. In particular, we analyze detailed facial features such as face shape (long face, angular chin, etc.) and forehead-eye ratio to recommend hairstyles that are in harmony with the user’s individual characteristics. The proposed algorithm effectively reflects the interaction between facial features and style elements that are missed in the existing methods by weighting the Jacquard similarity. Experimental results show that the proposed algorithm provides sophisticated recommendations with higher accuracy and user satisfaction than the traditional simple Jacquard similarity method. This research has the potential to contribute to increasing the level of personalized services in the beauty and fashion fields. |