Fuzzy Song Sets for Music Warehouses

The emergence of music recommendation systems calls for the development of new data management technologies able to query vast music collections. In this paper, we define fuzzy song sets and an algebra to manipulate them. We present a music warehouse prototype able to perform efficient nearest neighbor searches in an arbitrary song similarity space. Using fuzzy song sets, the music warehouse offers a practical solution to the all musical data management scenarios provided: song comparisons, user musical preferences and user feedback. We investigate three practical approaches to tackle the storage issues of fuzzy song sets: tables, arrays and bitmaps. Finally, we confront theoretical estimates to concrete implementation results and prove that, from a storage perspective, arrays and bitmaps are both effective data structure solutions.