PROJECT TITLE :

What Strikes the Strings of Your Heart?–Multi-Label Dimensionality Reduction for Music Emotion Analysis via Brain Imaging

ABSTRACT:

After twenty years of in depth study in psychology, some musical factors are identified which will evoke sure sorts of emotions. However, the underlying mechanism of the link between music and emotion remains unanswered. This paper intends to seek out the real correlates of music emotion by exploring a systematic and quantitative framework. The task is formulated as a dimensionality reduction downside, which seeks the whole and compact feature set with intrinsic correlates for the given objectives. Since a song generally elicits more than one emotion, we explore dimensionality reduction techniques for multi-label classification. One difficult downside is that the exhausting label cannot represent the extent of the emotion and it is additionally tough to ask the subjects to quantize their feelings. This work tries utilizing the electroencephalography (EEG) signal to unravel this challenge. A learning scheme called EEG-primarily based emotion smoothing ( ) and a bilinear multi-emotion similarity preserving embedding (BME-SPE) algorithm are proposed. We validate the effectiveness of the proposed framework on customary dataset CAL-500. Many influential correlates have been identified and also the classification via those correlates has achieved smart performance. We have a tendency to build a Chinese music dataset in line with the identified correlates and realize that the music from completely different cultures might share similar emotions.