Large-scale music tag recommendation with explicit multiple attributes

Abstract

Social tagging can provide rich semantic information for largescale retrieval in music discovery. Such collaborative intelligence, however, also generates a high degree of tags unhelpful to discovery, some of which obfuscate critical information. Towards addressing these shortcomings, tag recommendation for more robust music discovery is an emerging topic of significance for researchers. However, current methods do not consider diversity of music attributes, often using simple heuristics such as tag frequency for filtering out irrelevant tags. Music attributes encompass any number of perceived dimensions, for instance vocalness, genre, and instrumentation. Many of these are underrepresented by current tag recommenders. We propose a scheme for tag recommendation using Explicit Multiple Attributes based on tag semantic similarity and music content. In our approach, the attribute space is explicitly constrained at the outset to a set that minimizes semantic loss and tag noise, while ensuring attribute diversity. Once the user uploads or browses a song, the system recommends a list of relevant tags in each attribute independently. To the best of our knowledge, this is the first method to consider Explicit Multiple Attributes for tag recommendation. Our system is designed for large-scale deployment, on the order of millions of objects. For processing largescale music data sets, we design parallel algorithms based on the MapReduce framework to perform large-scale music content and social tag analysis, train a model, and compute tag similarity. We evaluate our tag recommendation system on CAL-500 and a largescale data set (N = 77, 448 songs) generated by crawling Youtube and Last.fm. Our results indicate that our proposed method is both effective for recommending attribute-diverse relevant tags and efficient at scalable processing.

A survey of music similarity and recommendation from music context data

“…By encouraging users to play such games, a large number of songs can be efficiently annotated with semantic descriptors. Another recent trend to alleviate the data sparsity problem and to allow fast indexing in a "semantic" space is automatic tagging and propagation of tags based on alternative data sources, foremost low-level audio features [Sordo et al 2007;Eck et al 2008;Kim et al 2009;Shen et al 2010;Zhao et al 2010]. …”

A Review of Ontology-Based Tag Recommendation Approaches

“…These modules maintain a list of indexed multiple attribute fuzzy music semantic vectors (MA‐FMSVs) and multiple attribute tag distance vectors (MA‐TDVs). Songs database is explored throughout offline processing …”

“…This approach is applicable to the task of other multimedia content analysis as well as tag suggestion. But the performance of the framework is not evaluated using unequal and bigger sized CEMA and SEMA attribute spaces …”

“…Songs database is explored throughout offline processing. 19 Parallel multiple attributes concept detector and parallel occurrence cooccurrence algorithms are used to produce MA-FMSVs and MA-TDVs for every song. The system suggests different attribute tags for a user-offered song throughout online processing.…”

Reconnaissance des émotions dans la musique

“…(Turnbull et al,2008); ce qui a généré 1700 annotations musicales décrivant 500 chansons occidentales populaires. Six attributs sont utilisés : Mood, Genre, Instrument, Song, Usage et Vocal (Zhao et al,2010). Ces tags furent générés sous des conditions de contrôle expérimental et sont considérés par la communauté comme faisant figure d'autorité.Ils ont ainsi pu concevoir un système capable de générer un descriptif pour une chanson au hasard à partir de leur base de données.…”