Harvesting and Structuring Social Data in Music Information Retrieval

摘要: An exponentially growing amount of music and sound resources are being shared by communities users on the Internet. Social media content can be found with different levels structuring, contributing might experts or non-experts domain. Harvesting structuring this information semantically would very useful in context-aware Music Information Retrieval (MIR). Until now, scant research field has taken advantage use formal knowledge representations process information. We propose a methodology that combines Media Mining, Knowledge Extraction Natural Language Processing techniques, to extract meaningful context from social data. By using extracted we aim improve retrieval, discovery annotation resources. define three scenarios test develop our methodology.