Application of a parametric supermatrix multi-task multimodal deep learning model based on electroencephalogram and peripheral physiological signal fusion in emotion recognition
Xue Li, Piqiang Gong, Chuantao Li, Xiaohan Ding, Fuming Chen · 2026
Read Application of a parametric supermatrix multi-task multimodal deep learning model based on electroencephalogram and peripheral physiological signal fusion in emotion recognition by Xue Li, Piqiang Gong, Chuantao Li, Xiaohan Ding, Fuming Chen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13473384.txt.
Description
Li and colleagues built in 2026 a lightweight brain-signal model that recognises emotions from EEG and body signals together. Their model fuses brain waves with peripheral signals such as heart and skin responses. Features are arranged on a head-grid by frequency band, and an attention step weights the bands. A compressed final layer, built by matrix decomposition, keeps the model small.
On the DEAP dataset the model reached about ninety-six percent accuracy for both pleasantness and arousal judgements. On the SEED dataset, with more electrodes, it reached about ninety-four percent for three-class emotion recognition. Fusion tests showed which body signals added most value. The authors say the design balances accuracy with low computation, suiting wearable or clinical devices. Multi-task learning, where one model judges several emotion aspects at once, shared the workload.
Catalog record
| CREATOR | Xue Li, Piqiang Gong, Chuantao Li, Xiaohan Ding, Fuming Chen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 78 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |