Combining Eye Movements and EEG to Enhance Emotion Recognition

Authors: Yifei Lu, Wei-Long Zheng, Binbin Li, Bao-Liang Lu

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiment results demonstrate that modality fusion could significantly improve emotion recognition accuracy in comparison with single modality. The best accuracy achieved by fuzzy integral fusion strategy is 87.59%, whereas the accuracies of solely using eye movements and EEG data are 77.80% and 78.51%, respectively.
Researcher Affiliation Academia Yifei Lu1, , Wei-Long Zheng1, , Binbin Li1, and Bao-Liang Lu1,2, 1Department of Computer Science and Engineering 2Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University, Shanghai, China {luyifei0715,weilong,libinbin,bllu}@sjtu.edu.cn
Pseudocode No The paper provides mathematical definitions but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper states: 'The dataset used in this paper will be freely available to the academic community via the website3.' with a link to a dataset, but it does not mention releasing the source code for the methodology.
Open Datasets Yes The dataset used in this paper will be freely available to the academic community via the website3. http://bcmi.sjtu.edu.cn/ seed/index.html
Dataset Splits No The paper states: 'For evaluation, we use the data from the first nine trials as training data and the data from remaining six trials as testing data in the whole experiment.' It defines training and testing splits but does not mention a separate validation split.
Hardware Specification No The paper mentions 'SMI ETG eye tracking glasses2' and 'ESI Neuro Scan System' for data recording, but does not provide details on the hardware used for running computational experiments (e.g., specific CPU/GPU models, memory).
Software Dependencies No The paper mentions methods like 'linear dynamic system (LDS)' and 'support vector machine with linear kernel' but does not specify any software names with version numbers.
Experiment Setup No The paper describes the choice of classifier (Support Vector Machine with linear kernel) and data splitting, but does not provide specific hyperparameters or system-level training settings like learning rate, batch size, or epochs.