• PALA POOJA RATNAM
• Dr.K.VENKATA RAO

In this modern world providing listening platform for music to users is critical task. An efficient source is to built a music recommendation system. These systems use recommendation methods by extracting data. Recommendation is mostly done on the meta data of the song but the availability of huge amount of existing genres and recommending a new song with no meta data to the user's cold start problem are the major difficulties. The following predicts a model in a content based approach for Music Recommendation System. In this paper the recommendation is done basing on user's MUSIC attributes of song. These five attributes namely: Mellow, Unpretentious, Sophisticated Intense and Contemporary (MUSIC). For the recommendation, analysis is done on the dataset with song features taken from music service (spotify) and its respective MUSIC attributes. Regression techniques-Support Vector Regression(SVR), Isotonic Regression(IR) are used for estimating MUSIC attributes and Classification technique(Random Forest) to examine a better Regression technique and cosine similarity to find similarity among users. The accuracy and root mean square error (RMSE)demonstrate the effectiveness of MUSIC model in music recommendation and also solve a cold start problem.

Music Recommendation System, Spotify, Isotonic Regression, Machine learning.

"A MUSIC RECOMMENDATION SYSTEM USING MUSIC ATTRIBUTES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.109-115, June 2019, Available :http://www.jetir.org/papers/JETIR1907J18.pdf

"A MUSIC RECOMMENDATION SYSTEM USING MUSIC ATTRIBUTES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp109-115, June 2019, Available at : http://www.jetir.org/papers/JETIR1907J18.pdf