Emotion Detection from Physiological Markers Using Machine Learning - Research Paper Summary

Aug 5, 2026

Uploaded by Eden

Emotion recognition in human-centered computer and robotic systems enables adaptive interactions that feel natural and supportive to users. This study experiments with machine learning-based emotion recognition from physiological markers using video stimuli that evoke distinct emotions. Heart rate (HR) and respiratory rate (RR) are analyzed for seven emotions: anger, sadness, fear, amusement, neutrality, surprise, and happiness/joy. Three classifiers—Random Forest, SVM, and J48—are compared, with J48 achieving the highest overall accuracy. RR most clearly distinguishes fear and sadness, while sadness is also best identified via HR, and observed gender differences further inform model interpretation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Emotion recognition in human-centered systems

Emotion definition and theories

Motivation for physiological-marker based detection

Potential application domains

Emotion Detection from Physiological Markers Using Machine Learning M. Kocaleva Vitanova*, A. Stojanova Ilievska, N. Koceska and S. Koceski Faculty of Computer Science, Goce Delcev University, Krste Misirkov 10A, Stip, Republic of North Macedonia [E-mail: mirjana.kocaleva@ugd.edu.mk](E-mail: mirjana.kocaleva@ugd.edu.mk), [aleksandra.ilievska@ugd.edu.mk](aleksandra.ilievska@ugd.edu.mk), [natasa.koceska@ugd.edu.mk](natasa.koceska@ugd.edu.mk), [saso.koceski@ugd.edu.mk](saso.koceski@ugd.edu.mk) *Corresponding author Keywords: emotion recognition, machine learning, physiological markers, heart rate, respiratory rate. Received: October 24, 2024 Human emotion recognition in computer and robotic systems is crucial because it allows these systems to respond to users in a way that feels natural and supportive. By interpreting emotional cues, these systems can adjust their interactions—offering empathy, encouragement, or even assistance during times of distress—enhancing user satisfaction and making technology more accessible and engaging. Emotion recognition methods include analyzing facial expressions, vocal tone, and physiological signals, with the latter being especially effective because physiological data offers objective, real-time insights that are less susceptible to misinterpretation or masking than visible expressions. In this paper, we conducted an experiment for emotion recognition from physiological markers using machine learning algorithms. Each of the participants involved in the experiment was exposed to video stimuli designed to evoke specific emotions. Using physiological markers such as heart rate (HR) and respiratory rate (RR), seven emotions—anger, sadness, fear, amusement, neutrality, surprise, and happiness/joy—were analyzed. Three classification methods Random-forest, SVMandJ48 were used. According to the results from the experimental evaluation, the highest accuracy for classifying emotions, based on both HR andRR across all emotions, was obtained with J48 algorithm. Specifically, the most clearly expressed and distinguishable emotions through RR were fear and sadness, with classification accuracies of 96.43% and 92.86%, respectively. Sadness was also the most accurately classified emotion through HR, with an accuracy of 85. 71%. Gender differences were noted, with females reacting more to sadness and males to happiness. Povzetek: Analiziranoje strojno prepoznavanje čustev iz fizioloških markerjev, kot sta srčni utrip in hitrostdihanja. Največjo kvaliteto pri prepoznavanju čustev je dosegel algoritem J48, z največjo ntočnostjo pri strahu in žalosti. 1 Introduction There has been a long-standing academic debate about the definition of emotion, without any widely agreed and universally accepted definition due to its complexity and diverse perspectives in psychology, neurology, and philosophy. However, contemporary study of emotions is usually based on the James-Lange theory of emotions, which suggests that a perceived stimulus is triggering physiological responses that are consequently felt as an emotion [1] . Emotions could be positive like love, happiness, hope, joy, affection, gratitude or negative such as anxiety, jealousy, frustration etc. Our own emotions can help us make sense of every situation. For example, appraisal theory states that our emotions are accompanied by inferences about the situation or environment in which we find ourselves [2], [3] . For a long period of time, emotions have been attributed to living beings or have been observed as unique human traits because only humans can experience a full range of emotions that are intrinsic to human experience, influencing cognition, behavior, decision-making, and social interactions. Nevertheless, throughout history, there have been various ideas and aspirations to create machines or artificial entities capable of emotions or human-like behaviors. This concept can be traced back to ancient myths, literature, and philosophical

FAQs

Why is emotion recognition important for computer and robotic systems?

It lets systems interpret emotional cues and adapt interactions to feel more natural, supportive, and engaging, improving user satisfaction and accessibility.

Which physiological markers and emotions are analyzed in this study?

The study uses heart rate (HR) and respiratory rate (RR) to analyze seven emotions: anger, sadness, fear, amusement, neutrality, surprise, and happiness/joy.

Which machine learning method performs best, and what are the key classification findings?

J48 achieves the highest overall accuracy. Fear and sadness are most distinguishable using RR, and sadness is most accurately classified using HR; gender differences are also reported.

Aug 5, 2026

Uploaded by Eden