ANALYSIS OF MUSIC CONTENT RECOMMENDATION SYSTEM USING AGLOMERATIVE HIERARCHICAL CLUSTERING METHOD IN SVARA PLATFORM
Nisa Hanum Harani(a*), Wahyu Marutie Adjie(a) , Anny Nurbasari(b)
(a)Politeknik Pos Indonesia
Jalan Sariasih No.54, Sarijadi, Sukasari, Kota Bandung, Jawa Barat 40151
*nisahanum[at]poltekpos.ac.id
(b)Universitas Kristen Maranatha
M.P.H, Jl. Surya Sumantri No.65, Sukawarna, Sukajadi, Kota Bandung, Jawa Barat
Abstract
SVARA is an internet radio application developed for music and radio fans that provides music, radio, podcast and social media content. The existence of a recommendation system will help recommend music content to listeners based on the mood of the listeners. To display music predictions that fit the listeners mood is by grouping music with similar attributes. This research begins with the random sampling stage of music data on the SVARA server, then the data pre-processing stage to get the value of features in the music data with feature extraction using the python pyAudioAnalysis library. The features obtained from the results of extracting music data become attributes for doing music grouping on moods. Music data grouping uses the DataMining method with the Agglomerative Hierarchical Clustering algorithm in 50 samples of music data. Mood parameters consisted of joy, relax, anxiety and depression. The application of the Hierarchical Agglomerative algorithm makes music data in the SVARA application grouped according to mood.
Keywords: Recommendation Systems;Python, pyAudioAnalysis, Data Mining, Clustering, Agglomerative Hierarchical
Topic: Informatic and Information System