[C-15] Electronics Simulation Technology

11:15 AM - 11:30 AM JST(2:15 AM - 2:30 AM UTC)Candidate for Academic Incentive Award

[C-15-21] Investigation of Emotion Classification from Musical Pieces Considering Tone Changes

◎Yuki Furuhashi¹, Seiya Kishimoto¹, Shinichiro Ohnuki¹ (1. Nihon Univ.)

Keywords:

Emotion classification,MIDI,Machine Learning,Random Forest,Tone Changes

Emotion classification from music holds great promise for applications in various fields, including music retrieval. While a variety of approaches have been proposed in previous studies, challenges remain in addressing emotion classification that accounts for tonal changes within a piece. In this report, we extract structural musical features from MIDI (Musical Instrument Digital Interface) data and examine a three-category emotion classification—"anger," "sadness," and "joy"—that takes these tonal changes into consideration.