China
Africa
Explore chapters and articles related to this topic
EEG-based Deep Emotional Diagnosis: A Comparative Study
Published in Richard Jiang, Li Zhang, Hua-Liang Wei, Danny Crookes, Paul Chazot, Recent Advances in AI-enabled Automated Medical Diagnosis, 2022
Geyi Liu, Zhaonian Zhang, Richard Jiang, Danny Crookes, Paul Chazot
Emotion plays an important role in human decision, interaction and cognition [1]. Due to the rapid development of society and the accelerated pace of life, people often feel pressure and anxiety. The persistence of this situation may lead to a variety of health problems or depression, thereby affecting people’s daily life and self-development. Therefore, emotion recognition is gradually becoming a practical topic of researchers. Nowadays, emotion recognition is used in many fields, such as text, speech, expression, and posture. But these methods are subjective and cannot guarantee the authenticity of emotion. Physiological and psychological studies show that the changes of physiological signals are often closer to people’s real emotions than facial expression, posture or voice [2]. Electroencephalogram (EEG) reflects all kinds of electrical activities and functional states of the brain, and contains the effective information of human emotional state. Using EEG signals for emotion detection has more advantages than other methods [3]. EEG based on emotion recognition will provide an accurate emotion in many fields.
Emotion Detection System
Published in Shampa Sen, Leonid Datta, Sayak Mitra, Machine Learning and IoT, 2018
Adrish Bhattacharya, Vibhash Chandra, Leonid Datta
Emotion recognition systems have the ability to humanize digital interactions by creating artificial emotional intelligence in machines. Since our interactions with technology are becoming increasingly conversational, artificial emotional intelligence has become a critical component of today's technology. Although a lot of work is going in this field of study and some amazing breakthroughs have been made, we have only scratched the tip of the iceberg. A lot of questions still remain unanswered and newer challenges crop up as we delve deeper into this field of study. However, these systems have the potential of disrupting industries from health care to education, and anything in between. Emotion recognition systems will not only revolutionize human interactions with machines but also allow machines to be more human.
Analysis of RNNs and Different ML and DL Classifiers on Speech-Based Emotion Recognition System Using Linear and Nonlinear Features
Published in Amit Kumar Tyagi, Ajith Abraham, Recurrent Neural Networks, 2023
Shivesh Jha, Sanay Shah, Raj Ghamsani, Preet Sanghavi, Narendra M. Shekokar
With high computing power and the advancement in the field of ML and deep learning, machines are able to achieve a high benchmark in performing tasks where human action or human intelligence is required. Communication with machines is getting popular these days; digital assistants like Alexa, Siri, Google Assistant, etc. are becoming ubiquitous. But none of these truly recognize our emotion and none of them react or respond in a way humans would do. There is a need for emotion recognition for machines to behave, act, and make human-like decisions. This is why the need for identifying human emotions has gained attention of many researchers.
AI’s Humanoid Appearance Can Affect Human Perceptions of Its Emotional Capability: Evidence from Self-Reported Data in the U.S
Published in International Journal of Human–Computer Interaction, 2023
Quan-Hoang Vuong, Viet-Phuong La, Minh-Hoang Nguyen, Ruining Jin, Minh-Khanh La, Tam-Tri Le
The research field of emotional AI has contributed to the development of emotion recognition technologies that facilitate the interactions between us and machines (Mantello et al., 2023; McStay, 2020). While such efforts are valuable within the current technological explosion, a conceptual concern should be addressed. The human mind is a multiplex information processing system that has yet to be understood clearly. Moreover, the present landscape of theoretical investigation into the working of simpler “minds,” such as in other biological systems or machines, also faces obstacles (Damasio, 2005). Debates surrounding how humans perceive AI’s capabilities of emotion go in various directions (Katz, 2012; Nader et al., 2022; Russell & Norvig, 2009). Suppose we are still quite unsure about the nature and functions of our own emotions and how we perceive the “emotion” of simpler systems. In that case, it probably is extremely challenging to create accurate and efficient AI (a simpler processing system compared to the human mind) that is tasked with assessing human emotions.
Eye Tracking, Usability, and User Experience: A Systematic Review
Published in International Journal of Human–Computer Interaction, 2023
Jakub Štěpán Novák, Jan Masner, Petr Benda, Pavel Šimek, Vojtěch Merunka
Melih et al. (2021) went beyond base eye movement segmentation. In their research, machine learning approaches, particularly logistic regression, support vector machines, and random forests were used to evaluate users’ familiarity with tested web applications yielding an accuracy of results of about 72% for datasets based on raw data analysis (Melih et al., 2021). A similar accuracy has been achieved in a study that used a neural network for emotion recognition, yielding a 70% accuracy rate (Filko & Martinovíc, 2013). Machine learning can also be used for automated gaze data mapping into relevant Areas Of Interest (AOIs) (Wolf et al., 2018). However, the authors noted that a fully automated approach is not seamlessly possible for practical experiments in real-life scenarios because of the constant need for manual assignment of gaze points frame by frame to reference datasets.
Ethical Awareness in Paralinguistics: A Taxonomy of Applications
Published in International Journal of Human–Computer Interaction, 2023
Anton Batliner, Michael Neumann, Felix Burkhardt, Alice Baird, Sarina Meyer, Ngoc Thang Vu, Björn W. Schuller
There appears to be a significant difference between corporate responsibility and genuine ethical awareness (Benjamins et al., 2019). In other words, as yet, it seems that ethics in areas of AI are instead a set of buzz words to build commercial trust, show intention, or address a specific responsibility rather than low-level sincere concern. This becomes evident in cases of conflicts between employees and companies10 (Ebell et al., 2021). Despite this, there have been regulations at a higher governmental level in recent time that may encourage a faster change, and foster a more honest intention. For example, the European Commission presented their guidelines for Artificial Intelligence11 in which they suggest the explicit banning of manipulative AI systems which may implement aspects of social scoring, limiting biometric-based real-time remote identification, as well as “special transparency requirements on all emotion recognition and biometric categorisation systems.”