Recognition of Negative Emotion using Long Short-Term Memory with Bio-Signal Feature Compression.

Jeeeun Lee,Sun K Yoo

doi:10.3390/s20020573

Abstract

Negative emotion is one reason why stress causes negative feedback. Therefore, many studies are being done to recognize negative emotions. However, emotion is difficult to classify because it is subjective and difficult to quantify. Moreover, emotion changes over time and is affected by mood. Therefore, we measured electrocardiogram (ECG), skin temperature (ST), and galvanic skin response (GSR) to detect objective indicators. We also compressed the features associated with emotion using a stacked auto-encoder (SAE). Finally, the compressed features and time information were used in training through long short-term memory (LSTM). As a result, the proposed LSTM used with the feature compression model showed the highest accuracy (99.4%) for recognizing negative emotions. The results of the suggested model were 11.3% higher than with a neural network (NN) and 5.6% higher than with SAE.

Highlights

Emotion occurs through a complex interaction of stimuli and is used as an indicator to infer one’s psychological and emotional state [1]
This study study utilized utilized a stacked a stacked auto-encoder auto-encoder (SAE)
In terms of the mean and standard deviation, the mean heart rate variability (HRV), standard deviation of R-peak intervals (SDNN), RMSSD, and pNN50 extracted from ECG, the SD skin temperature (ST) extracted from total frequency power (TF) and ST, and the features extracted from galvanic skin response (GSR) were different depending on emotions

Summary

Introduction

Emotion occurs through a complex interaction of stimuli and is used as an indicator to infer one’s psychological and emotional state [1]. Recognizing a negative emotion is the starting point for addressing risk factors. From this perspective, it is very important to classify negative emotions. A variety of methods including questionnaire evaluation interviews, facial expressions, and gestures are used to discriminate emotions [3]. These techniques reflect personal thinking, culture, age, and gender and can result in manipulation [4,5,6,7]. Bio-signals do not allow any intentional manipulation, they present one’s personal psychological state according to stimuli. If a bio-signal is used to discriminate emotions, it is possible to obtain more objective and more accurate information than with personal responses

Results

Discussion

Conclusion