Emotion Recognition Using Biological Signal in Intelligent Space

Abstract

In this study, we focus on emotion recognition for service robots in the living space based on Electrocardiogram (ECG). An emotional state is important information that allows a robot system to provide appropriate services in way that are more in tune with users' needs and preferences. Moreover, the users' emotional state can be feedbacks to evaluate user's level of satisfaction in the services. We apply a diagnosis method that uses both interbeat and within-beat features of ECG. The post hoc tests in Analysis of Variance (ANOVA) showed that our approach satisfies more confidence level of difference between emotions than conventional methods. Our system design was based on wireless and wearable biological sensor for mobility and convenience of users' daily lifestyle.

Electrocardiogram‐based emotion recognition system using empirical mode decomposition and discrete Fourier transform

“…Table shows the classification rates for the normal power, normal entropy, spectral power and spectral entropy parameters obtained using the Hilbert transform and discrete Fourier transform on the IMFs. The classification was carried out in a subject‐independent way for all the 30 subjects, in contrast to the previous works, most of which were subject dependent (Wan‐Hui et al ., ; Maaoui and Pruski, ; Rattanyu et al ., ).…”

“…The accuracy obtained in this work is better than that of the works by previous researchers in classifying six classes of emotions (Maaoui and Pruski, ; Rattanyu et al ., ). In the work performed by Rattanyu et al ., using analysis of variance, the classification accuracy is 61% for classifying six emotional states.…”

Emotion Recognition Using Wearables: A Systematic Literature Review - Work-in-progress

“…Many authors decided to apply their own emotion categories, usually as a subset or small modification of the classical ones, e.g. happy-sad-anger + neutral state [19] extended in [20] with fear, happy-sad-anger-pain [21] joysadness-anger-pleasure [22], or anger-sadness-fear-disgustjoy/amusement + neutral [23], also extended with the seventh one -affection [24], anger-sadness-fear-surprise-frustrationamusement [25], extended in [26] with disgust-other (they did not applied these two to reasoning), excited-bored-stressedrelaxed-happy [27]. The number of reported categorical emotions can be even greater like as many as eight+neutral in [28].…”

“…Used by ECG [18], [20], [22], [23], [30], [36] PPG [24], [27] GSR, EDA, EDR, SC [5], [7], [15], [17], [21], [24]- [27], [29], [31]- [34], [36], [39] EEG [19], [28], [35]-[38] RSP…”

Classification of emotional states from electrocardiogram signals: a non-linear approach based on hurst

“…QRS complex, an activity of the ANS that is derived from ECG provides information of the myocardial conduction system and can be used to understand the emotions experienced by a person [13]. This research being in the infancy stages, the performance of the different systems developed to recognize human emotions varies from 37% to 100%, depending on factors such as the number of emotions, the number of subjects, type of processing and types of emotion elicitation [2,3,14,15]. Some of the works are done in a subject dependent way where the training and testing data belong to the same subject [4,15].…”