On Evaluation of Inter- and Intra-Rater Agreement in Music Recommendation

Flexer, Arthur; Lallai, Taric; Rašl, Katja

doi:10.5334/tismir.107

Cited by 25 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Particularly, the proposed HC and MIX methods appear to significantly outperform the random baseline for SGD and XGB models. This finding leads us to promote the creation of subjectivity-aware machine learning methods which could have a high impact in novel applications of immersion in virtual reality (Warp et al, 2022) and emotion-based music recommendation (Grekow, 2021;Tarnowska, 2021)-several other tasks display low inter-rater agreement too: music auto-tagging (Bigand & Aucouturier, 2013), music similarity and diversity (Flexer et al, 2021;Porcaro et al, 2022), automatic chord estimation (Koops et al, 2019), and beat tracking (Holzapfel et al, 2012). In short, MER has been openly criticized due to the subjectivity issue (Gómez-Cañónet al, 2021) -however we advocate for "embracing subjectivity and potentially leveraging the opportunities it offers for better learning" (Rizos & Schuller, 2020).…”

Section: Discussionmentioning

confidence: 99%

TROMPA-MER: an open dataset for personalized music emotion recognition

et al. 2022

View full text Add to dashboard Cite

We present a platform and a dataset to help research on Music Emotion Recognition (MER). We developed the Music Enthusiasts platform aiming to improve the gathering and analysis of the so-called “ground truth” needed as input to MER systems. Firstly, our platform involves engaging participants using citizen science strategies and generate music emotion annotations – the platform presents didactic information and musical recommendations as incentivization, and collects data regarding demographics, mood, and language from each participant. Participants annotated each music excerpt with single free-text emotion words (in native language), distinct forced-choice emotion categories, preference, and familiarity. Additionally, participants stated the reasons for each annotation – including those distinctive of emotion perception and emotion induction. Secondly, our dataset was created for personalized MER and contains information from 181 participants, 4721 annotations, and 1161 music excerpts. To showcase the use of the dataset, we present a methodology for personalization of MER models based on active learning. The experiments show evidence that using the judgment of the crowd as prior knowledge for active learning allows for more effective personalization of MER systems for this particular dataset. Our dataset is publicly available and we invite researchers to use it for testing MER systems.

show abstract

Section: Discussionmentioning

confidence: 99%

TROMPA-MER: an open dataset for personalized music emotion recognition

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Particularly, the proposed HC and MIX methods appear to significantly outperform the random baseline for SGD and XGB models. This finding leads us to promote the creation of subjectivityaware machine learning methods -several other tasks have low inter-rater agreement too (music auto-tagging [65], music similarity and diversity [66,67], automatic chord estimation [68], and beat tracking [69]). In short, MER has been openly criticized due to the subjectivity issue [6] -however we advocate for "embracing subjectivity and potentially leveraging the opportunities it offers for better learning" [70].…”

Section: Discussionmentioning

confidence: 99%

TROMPA-MER: an open dataset for personalized Music Emotion Recognition

Sebastián

Gutiérrez-Páez

Porcaro

et al. 2022

Preprint

View full text Add to dashboard Cite

We present a platform and a dataset to help research on Music Emotion Recognition (MER). We developed the Music Enthusiasts platform aiming to improve the gathering and analysis of the so-called “ground truth” needed as input to MER systems. Firstly, our platform involves engaging participants using citizen science strategies and generate music emotion annotations -- the platform presents didactic information and musical recommendations as incentivization, and collects data regarding demographics, mood, and language from each participant. Participants annotated each music excerpt with single free-text emotion words (in native language), distinct forced-choice emotion categories, preference, and familiarity. Additionally, participants stated the reasons for each annotation -- including those distinctive of emotion perception and emotion induction. Secondly, our dataset was created for personalized MER and contains information from 181 participants, 4721 annotations, and 1161 music excerpts. To showcase the use of the dataset, we present a methodology for personalization of MER models based on active learning. The experiments show evidence that using the judgment of the crowd as prior knowledge for active learning allows for more effective personalization of MER systems for this particular dataset. Our dataset is publicly available and we invite researchers to use it for testing MER systems.

show abstract

“…The scale yields measures of moods including pleasant-unpleasant mood, arousal-calm mood, as well as scores for positive-tired and negative-calm mood (Mayer & Gaschke, 1988). The scale has been validated in numerous studies, including Aldrich et al (2021), Flexer et al (2021), andNugraha et al (2020). In order to evaluate any possible correlation between subjective assessment and quantitatively gathered data, participants filled out a BMIS form after each condition.…”

Section: Methodsmentioning

confidence: 99%

Binaural Beats and Music: A Preliminary Case Report

Elizabeth Krasnoff,

Chevalier

2024

JSE

View full text Add to dashboard Cite

We recruited subjects who reported feeling stressed and needing a break and were seeking an opportunity to experience relaxation. Inaudible binaural beats (BB) were used to measure the ability of BB to induce relaxation. Measurements included EEG, Menlascan (cardiovascular and ANS) (Menla Technologies, Independence, Missouri), and GDV (Gas Discharge Visualization/Kirlian Photography) (Korotkov et al., 2020). Further, a Big Five Character Assessment was administered, and individual Brief Mood Introspection Scales (BMIS) were collected for each condition. Subjects experienced an improvement in brain function, microcirculation or cardiovascular score, bioenergy, and a calmer brain after adding binaural beats to brown noise or to music plus brown noise, although the Menlascan scores and Big Five character assessment results were less conclusive. BB seems to have profound effects on the physiology of subjects, and since the beats were not audible, these effects could not be attributed to the placebo effect. These results are encouraging in terms of developing musical products incorporating BB to affect human neural rhythms and corollary states of consciousness and warrant further research with more subjects and different frequencies of BB and different music tracks.

show abstract

On Evaluation of Inter- and Intra-Rater Agreement in Music Recommendation

Cited by 25 publications

References 27 publications

TROMPA-MER: an open dataset for personalized music emotion recognition

TROMPA-MER: an open dataset for personalized music emotion recognition

TROMPA-MER: an open dataset for personalized Music Emotion Recognition

Binaural Beats and Music: A Preliminary Case Report

Contact Info

Product

Resources

About