A tag-based hybrid music recommendation system using semantic relations and multi-domain information
11th IEEE International Conference on Data Mining Workshops, ICDMW 2011, Vancouver, Kanada, 11 Aralık 2011, ss.548-554, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/icdmw.2011.17
- Basıldığı Şehir: Vancouver
- Basıldığı Ülke: Kanada
- Sayfa Sayıları: ss.548-554
- Anahtar Kelimeler: Dimensionality reduction, Recommendation systems, Semantic relations, Social tagging, User profiling
- TED Üniversitesi Adresli: Hayır
Özet
In this paper, we propose a hybrid approach for music recommendation. Firstly, we describe an approach for creating music recommendations based on user-supplied tags that are augmented with a hierarchical structure extracted for top level genres from Dbpedia. In this structure, each genre is represented by its stylistic origins, typical instruments, derivative forms, sub genres and fusion genres. We use this well-organized structure in dimensionality reduction in user and item profiling. We compare two recommenders; one using our method and the other using Latent Semantic Analysis (LSA) in dimensionality reduction. The recommender using our approach outperforms the other. In addition to different dimensionality reduction methods, we evaluate the recommenders with different user profiling methods. Moreover, our approach collects personal interests (favorite movies and television series) from the Facebook profiles. These user profiles are then used to find the similarity between users. At the end, items belonging to the most similar users' profiles and having a high score against users' profiles are recommended. Thus, we have focused on a hybrid system using tag-based contextual information of music tracks and user interests acquired from Facebook profiles. Initial results are promising such that using similarities of users affects the recommendation positively. © 2011 IEEE.