Multi-Label Classification of Music into Emotions

By Konstantinos Trohidis, Grigorios Tsoumakas, George M. Kalliris, Ioannis P. Vlahavas. International Society for Music Information Retrieval Conference, 2008

811 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/0caf7892917a53011e062c9f9057551a601b6695

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Online Metric Learning for Multi-Label Classification (2020, methodology): “To evaluate the performance of our proposed online metric learning algorithm, we conduct experiments on four benchmark datasets: Corel5k, Enron, Medical and Emotions.”
A Survey of Audio-Based Music Classification and Annotation (2011, methodology): “K-NN [116] and SVM [117] are the two most popular classifiers used for both general classification problems and in music classification as well.”
Explainable and Interpretable Models in Computer Vision and Machine Learning (2017, methodology): “Lewis 1992, 2004), but nowadays multi-label methods are used in applications as diverse as music categorization (Trohidis et al. 2008), semantic scene classification (Boutell et al.”
A Literature Survey on Algorithms for Multi-label Learning (2010, background): “This music emotions dataset contains 593 such music instances and each belonging to any of 6 classes [51].”
A Tutorial on Multilabel Learning (2015, methodology): “—Detection of emotions in music is an MLL problem in which songs are classified simultaneously in several categories [Trohidis et al. 2008; Ma et al. 2009].”
Large-Scale Multi-label Text Classification - Revisiting Neural Networks (2013, background): “In the literature, one can find a number of multi-label classification approaches for a variety of tasks in different domains such as bioinformatics [1], music [33], and text [8].”
On label dependence and loss minimization in multi-label classification (2012, methodology): “The EMOTIONS data was created from a selection of songs from 233 musical albums (Trohidis et al. 2008).”
Multilabel Classification with Principal Label Space Transformation (2012, background): “, 2006), music classification (Trohidis et al., 2008) and text categorization (Schapire and Singer, 2000), multi-label classification is attracting more and more research attention.”
Feature selection for multi-label naive Bayes classification (2009, methodology): “Furthermore, multi-label learning has also been applied to data mining tasks such as association rule mining [30,37,41] and music information analysis [38].”
Combining instance-based learning and logistic regression for multilabel classification (2009, methodology): “The emotions data was created from a selection of songs from 233 musical albums [18].”
Multi-label Classification via Feature-aware Implicit Label Space Encoding (2014, methodology): “…problem with many classes (i.e. a large vocabulary), many effective methods were proposed, like constructing a hierarchy of multi-label classifiers (Tsoumakas et al., 2008a), refining the output of heuristic efficient classifiers (Dekel & Shamir, 2010), or performing label selection…”
Training deep networks for facial expression recognition with crowd-sourced label distribution (2016, methodology): “In this paper, we adopt the latest deep convolutional neural networks (DCNN) architecture, and evaluate the effectiveness of four different schemes to train emotion recognition on crowd-sourced labels.”
Two-Stage Label Embedding via Neural Factorization Machine for Multi-Label Classification (2019, background): “• Emotions (Trohidis et al. 2008): A music dataset consisting of hundreds of songs from 6 genres.”
The ordinal nature of emotions (2017, methodology): “in affective computing have used PL for affect detection and retrieval through images [42], [43], videos [39], [44], music [31], [45], sounds [38], speech [27], [37], [46], games [27], [47], [48] and text [49].”
Inverse random under sampling for class imbalance problem and its application to multi-label classification (2012, background): “The collection was created from 233 musical albums choosing three songs from each album [47].”
Lift: Multi-Label Learning with Label-Specific Features (2011, background): “…and related communities, which has been successfully applied to diverse applications such as bioinformatics [1], [5], multimedia contents analysis including image [3], [4], [49], audio [27], [34], [42], and video [35], [41], [48], web mining [25], [33], [36], information retrieval [52], [54], etc.”
Classifier chains for multi-label classification (2009, methodology): “MLkNN performs well against BM, but in [12], it did not perform as well as RAKEL (see below), which we use in our evaluation.”