It's Only Words And Words Are All I Have
원문 초록
The central idea of this paper is to demonstrate the strength of lyrics for music mining and natural language processing (NLP) tasks using the distributed representation paradigm. For music mining, we address two prediction tasks for songs: genre and popularity. Existing works for both these problems have two major bottlenecks. First, they represent lyrics using handcrafted features that require intricate knowledge of language and music. Second, they consider lyrics as a weak indicator of genre and popularity. We overcome both the bottlenecks by representing lyrics using distributed representation. In our work, genre identification is a multi-class classification task whereas popularity prediction is a binary classification task. We achieve an F1 score of around 0.6 for both the tasks using only lyrics. Distributed representation of words is now heavily used for various NLP algorithms. We show that lyrics can be used to improve the quality of this representation.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
VenusX: Unlocking Fine-Grained Functional Understanding of Proteins
Detecting immune cells with label-free two-photon autofluorescence and deep learning
Leveraging Cascaded Binary Classification and Multimodal Fusion for Dementia Detection through Spontaneous Speech
Self-Classification Enhancement and Correction for Weakly Supervised Object Detection
In-Context Learning for Label-Efficient Cancer Image Classification in Oncology
LLSF - Learning Label Specific Features for Multi-Label Classifcation
토론
이 논문과 연결된 포럼 글이 아직 없습니다.