Titulo Estágio
Metaphor and Irony Mining: Advanced Techniques for Emotion Recognition in Song Lyrics
Local do Estágio
DEI-FCTUC
Enquadramento
The area of Music Emotion Recognition (MER), a sub-area of Music Information Retrieval (MIR), has received increasing interest from the international scientific community, both academic and industrial. In fact, companies such as Spotify, Pandora, Google, Sony, Philips Research, Gracenote, among many others, have invested heavily in the area in recent years and use this aspect in their applications.
The recognition of music in emotions contemplates the recognition of emotions in the two dimensions of music (audio and song lyrics). This is because the emotional perception that a listener has of a song can depend on any of these two dimensions with variable weights, always depending on ultimately the listener and the music.
The classic process of building prediction models for MER begins by extracting features from the audio and lyrics separately. For example, in audio, we have features related to the detection of rhythm, melody, tempo, etc., and for lyric features, we have semantic, structural, stylistic, and content-based features. Then, the process continues with the construction of predictive models that use these features.
It is currently clear that the effectiveness of the models built tends to stabilize if the so-called glass ceiling is not broken, i.e., if more dedicated features from an emotional point of view are not used.
Features related to the semantic analysis of the text, in this case, song lyrics, could be an important contribution to breaking this glass ceiling and obtaining more emotionally important features.
The objective of this project is related to the process of identifying figures of speech such as metaphors, sarcasm, and irony with the aim of building features from them that allow the implementation of more effective predictive models. The process involves the use of natural language processing techniques, text mining, feature engineering, and knowledge of applied machine learning.
Objetivo
The objectives to be achieved in this project are the following:
1. Research and development of linguistic analysis techniques, including metaphor and irony identification for static Lyrics-based MER (LMER)
2. Research and development of NLP and feature engineering approaches for static Lyrics-based MER (LMER) (following point 1)
Plano de Trabalhos - Semestre 1
1. State-of-the-art review: MER, figures of speech (metaphors, irony), NLP and feature engineering, machine learning for text in general and LMER in particular
2. Evaluation of some of the current algorithms (based on classical Natural Language Processing techniques and machine learning and following point 1) on the novel datasets developed within the research group
3. Writing of the Intermediate Report
Plano de Trabalhos - Semestre 2
4. Research and development of NLP and feature engineering approaches for static LMER (namely semantic or linguistic features)
5. Creation of predictive models using the created features
6. Critical evaluation of the developed models
7. Writing of a scientific paper and thesis
Condições
- Access to a toolbox for LMER (with feature extraction and deep learning modules), developed by our team
- Access to an LMER database, developed by our team
- Server access (hosted at DEI) with 10 high-performance GPU cards
Orientador
Rui Pedro Paiva, Ricardo Malheiro e Renato Panda
ruipedro@dei.uc.pt 📩