Recognizing Musical Entities in User-generated Content
This work addresses a gap in Music Information Retrieval by focusing on user-generated content, which is incremental as it adapts existing methods to a new, noisy domain.
The paper tackled the problem of recognizing musical entities in noisy user-generated content from Twitter, specifically for a classical music radio channel, by developing a method that combines formal radio schedules and tweets to improve recognition, achieving enhanced results through joint consideration of both data sources.
Recognizing Musical Entities is important for Music Information Retrieval (MIR) since it can improve the performance of several tasks such as music recommendation, genre classification or artist similarity. However, most entity recognition systems in the music domain have concentrated on formal texts (e.g. artists' biographies, encyclopedic articles, etc.), ignoring rich and noisy user-generated content. In this work, we present a novel method to recognize musical entities in Twitter content generated by users following a classical music radio channel. Our approach takes advantage of both formal radio schedule and users' tweets to improve entity recognition. We instantiate several machine learning algorithms to perform entity recognition combining task-specific and corpus-based features. We also show how to improve recognition results by jointly considering formal and user-generated content