Emotion-Aware Music Recommendation
Emotion is one of the major factors for users to determine service preference. Especially online music streaming services are in trend-sensitive industry, hence largely affected by user’s experience and reputation. Conventional music streaming services provide users keywords-based search for music. Accordingly it strongly relies on user’s prior knowledge and experience. It often fails to expose non-expert users to the music that the users are not familiar with. In this paper, we suggest an emotion-aware music recommendation system that proposes songs and artists based on the mood of each user.
First, we infer user’s emotion using real-time weather information. Second, we classify songs and artists which are favorable in different weather conditions. To do so, we collect and combine daily chart of K-pop music and weather history data to find the music preference in different weather. It is used to recommend timely and favorable music to users after capturing their mood implcitly.
Moreover the emotion-aware music recommendation system is extensible to provide a personalized service by using user’s social media, heartbeat, time, location, and so on. We expect this would enrich user experience noticeably. Being aware of user’s emotion will enable broad areas of industry to provide intelligent services in a user-friendly way.
KeywordsEmotion-aware system Recommendation system Data mining
This research was supported by Advancement of Collage Education (2016) funded by Ministry of Education.
- 1.RIAJ Yearbook 2015: IFPI 2013, 2014. Global Sales of Recorded Music, p. 24. Recording Industry Association of Japan (2015)Google Scholar
- 2.Weather Underground. http://www.wunderground.com/history/
- 3.Bugs Music Chart. http://music.bugs.co.kr/chart?wl_ref=M_left_02_01
- 6.Chen, H.-C., Chen, A.L.P.: A music recommendation system based on music data grouping and user interests. In: Proceedings of the Tenth International Conference on Information and Knowledge Management. ACM (2001)Google Scholar
- 11.Takács, G., Tikk, D.: Alternating least squares for personalized ranking. In: Proceedings of the Sixth ACM Conference on Recommender Systems. ACM (2012)Google Scholar
- 12.Scikit-learn Library. http://scikit-learn.org/
- 13.MacQueen, J.: Some methods for classification and analysis of multivariate observations. In: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, vol. 1, no. 14 (1967)Google Scholar
- 14.Mckinney, W.: Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython. O’Reilly Media, Inc., Sebastopol (2012)Google Scholar