A Channel Selection Method for Emotion Recognition From EEG Based on Swarm-Intelligence Algorithms

Abstract

Increasing demand for human-computer interaction applications has escalated the need for automatic emotion recognition as emotions are essential for natural communication. There are various information sources that can be used for recognizing emotions, such as speech, facial expressions, body movements, and physiological signals. Among those physiological signals are more reliable for better affective communication with machines since they are almost impossible to control. Therefore, automatic emotion recognition from EEG signals has been a topic intensely investigated. Emotions are experiences that arise various cognitive functions observed in different frequency bands involving multiple brain areas and recognition from EEG with high accuracies is only possible with a large number of features extracted from the whole brain in various bands. Emotion regulation also requires integration of cognitive functions and thus functional connectivity between regions should also be considered. In this paper, we extract 736 features based on spectral power and phase-locking values. We particularly focus on finding salient features for emotion recognition using swarm-intelligence (SI) algorithms. We applied well-known classification algorithms for recognizing positive and negative emotions using the feature sets that are selected by these algorithms. Besides, features that are selected by all of them commonly are used as a new feature set. We report accuracies between 56.27% and 60.29% on the average; noting that by decreasing the feature size by 87.17% (from 736 to 94.40) an average accuracy of 60.01±8.93 was obtained with the random forest classifier. We also highlight the efficient electrode locations for emotion recognition. As a result, we define 11 channels as dominant and promising classification results are obtained.

Emotion recognition with reduced channels using CWT based EEG feature representation and a CNN classifier

“…As discussed in section 3.3, around the mean-scale in figure 5, twenty scales within 11 − 30 out of 64 scales are selected, and any contribution is rarely found in the 3D EER plot outside this range. We also analyze the choice of selecting a more narrow range around the mean scale, such as thirteen scales (15-27) and five scales (18)(19)(20)(21)(22). The classification performance (accuracy) for valence and arousal cases are shown in figure 11 considering six sample subjects.…”

“…The performance of emotion classification using the DEAP dataset is compared to that reported by other recent methods in table 7. Similar to some existing methods [7,16,17,20,22], subject-dependent performance is investigated. In [22], 11 dominant EEG channels are selected, where they used a one-second window frame, and in [5], emotion recognition accuracy for 10, 14, 18 and 32 EEG channels are considered separately with a 4s window frame.…”

“…Similar to some existing methods [7,16,17,20,22], subject-dependent performance is investigated. In [22], 11 dominant EEG channels are selected, where they used a one-second window frame, and in [5], emotion recognition accuracy for 10, 14, 18 and 32 EEG channels are considered separately with a 4s window frame. In [19], the feature cube is constructed using the power spectral density, and a 6 second time window is selected to partition the EEG data.…”

“…In [22], sophisticated feature selection methods like Particle Swarm Optimization, Cuckoo Search, Grey Wolf Optimizer and Dragonfly are utilized. After that, well-known classification algorithms such as SVM and k-NN classifiers are used to identify the most effective channels for the emotion recognition task.…”

A Survey on EEG-Based Solutions for Emotion Recognition With a Low Number of Channels

“…In [173], Particle Swarm Optimization (PSO, [174]), Cuckoo Search (CS, [175]), Grey Wolf Optimizer (GWO, [176]), and Dragonfly ( [177]) are adopted to select relevant features. The selected features are used to choose the most relevant channels for an emotion classification problem on the DEAP dataset with SVM and k-NN classifiers.…”