In MULTIMEDIA '06: Proceedings of the 14th annual ACM international conference on Multimedia, pages 911-920, New York, NY, USA, 2006. Real-time computerized annotation of pictures. International Symposium on Music Information Retrieval, 2009. Easy as cba: A simple probabilistic model for tagging music. MapReduce: simplified data processing on large clusters. In MM '08: Proceeding of the 16th ACM international conference on Multimedia, pages 737-740, New York, NY, USA, 2008. Sheepdog: group and tag recommendation for flickr photos by automatic search-based learning. Map-reduce for machine learning on multicore. Autotagger: A model for predicting social tags from acoustic features on large music databases. Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions. Our results indicate that our proposed method is both effective for recommending attribute-diverse relevant tags and efficient at scalable processing. We evaluate our tag recommendation system on CAL-500 and a large-scale data set ($N = 77,448$ songs) generated by crawling Youtube and Last.fm. For processing large-scale 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.
Our system is designed for large-scale deployment, on the order of millions of objects. To the best of our knowledge, this is the first method to consider Explicit Multiple Attributes for tag recommendation. Once the user uploads or browses a song, the system recommends a list of relevant tags in each attribute independently. 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. We propose a scheme for tag recommendation using Explicit Multiple Attributes based on tag semantic similarity and music content. Many of these are underrepresented by current tag recommenders. Music attributes encompass any number of perceived dimensions, for instance vocalness, genre, and instrumentation. However, current methods do not consider diversity of music attributes, often using simple heuristics such as tag frequency for filtering out irrelevant tags.
Towards addressing these shortcomings, tag recommendation for more robust music discovery is an emerging topic of significance for researchers. Such collaborative intelligence, however, also generates a high degree of tags unhelpful to discovery, some of which obfuscate critical information. Social tagging can provide rich semantic information for large-scale retrieval in music discovery.
0 kommentar(er)
