Emotional Artificial Intelligence Research Articles

Emotional Artificial Intelligence: Methodologies, Benefits, and Drawbacks

Abstract: This survey paper explores various research studies and papers related to Emotional Artificial Intelligence (Emotion AI). Emotion AI has gained significant attention in recent years due to its potential applications in diverse fields, such as mental health, workplace surveillance, education, and music generation. The paper provides an overview of the methodologies employed in these studies, along with the benefits and drawbacks associated with each approach. By examining the literature, this paper aims to shed light on the current state of Emotional AI and its implications for different domains

Legal incentives for innovations in the emotional AI domain: a carrot and stick approach?

Abstract Emotions strongly influence the human way of living and life experiences. In this context, Artificial Intelligence (AI) technologies are crucial to pushing developments further. Although emotional AI-driven innovations are welcome in our society, they might also have negative effects on the interdependence and autonomy of natural persons. Thus, they might be challenged by several legal provisions in the EU such as the General Data Protection Regulation (GDPR) and the draft AI Act. Yet these inventions require considerable investment, where legal incentives such as intellectual property rights (IPR) are crucial. Indeed, it is also important to secure certainty as to the legal and ethical acceptability of such innovations. This article looks at emotional AI to investigate the interlinkage between technological innovations, legal incentives and ethics, through the lenses of patent law and fundamental rights, in order to shed light over the challenges, limitations, but also opportunities for the protection, commercialization and exploitation of emotional AI-related inventions. Our research offers new scientific knowledge on the largely under-explored issue of legal incentives for emotional AI-related inventions in the European framework. It also provides companies and inventors with key points to consider in decision-making related to investments in and incentives for emotional AI-related innovations, also elaborating on suggestions for the European legislator and policymakers to better stimulate and promote emotional AI technology through regulation.

The Infopolitics of feeling: How race and disability are configured in Emotion Recognition Technology

In this article, we argue that facial emotion recognition technology (facial ERT) reproduces historical forms of pseudoscience based on the concept of quantifiable and unequally distributed emotional capacity. Drawing on Kyla Schuller’s Biopolitics of Feeling and Colin Koopman’s theory of infopower, we put forward the term ‘the infopolitics of feeling’ to describe how facial ERT encodes culturally ‘correct’ or normative forms of emotional expression that have historically been used to define and delineate what it means to be human. To make this argument, we provide a close reading of Girl Decoded, the autobiography of Rana el Kaliouby, the founder and former CEO of the leading Emotion artificial intelligence (AI) firm Affectiva. Girl Decoded, we argue pits el Kaliouby herself – portrayed as the empathetic, liberal, emotionally expressive and ideal ‘feeling’ subject – against two non-normative figures: the unfeeling autist and the inscrutable Oriental who must be ‘cured’ through Affectiva’s facial ERT.

An Emotional Artificial Intelligence Based Attendance System for Class Room

Attendance through a camera using Artificial Intelligence (AI) and Deep Learning (DL) is a mod-ern approach to monitoring and tracking attendance in various settings such as schools, offices, and man-ufacturing facilities. It uses computer vision techniques and deep learning algorithms to automatically detect and identify individuals in an image or video captured by the camera. The process typically in-volves training the system on a dataset of labelled images of the individuals who will be taking attend-ance. This training data can include images of the individual's face, iris, or fingerprints, depending on the specific approach used. Once the system is trained, it can then use this knowledge to recognize these in-dividuals in new images captured by the camera. When an individual is recognized, the system can log their attendance in a database or other record-keeping system. This can be done in real-time, allowing for immediate tracking of attendance, or it can be done at a later time for batch processing. However, AI has been expanding its horizon and face recognition with deep learning techniques can be augmented with emotion recognition as well. Sometimes, students feel really sad and overwhelmed by their school work and other responsibilities. They might feel like they can't keep up or that they are not good enough. When these feelings last for a long time, it's called depression. Depression can make it hard for students to do their school work, be with friends and family, or even get out of bed in the morning. When students are feeling really sad and hopeless, they might think about hurting themselves or ending their lives. This is called suicide. It's important to know that suicide is preventable and there are people who can help. It's important to take care of our mental health, just like we take care of our physical health. If you ever feel sad or overwhelmed, it's important to talk to someone you trust and get help. One way to prevent suicide is to detect if someone is feeling sad and hopeless, which can be a sign of depression. One way to do this is by using a CCTV camera and AI and DL technology to analyze the person's facial expression, body language, and speech patterns. Emotional AI and DL can be a powerful tool in detecting depression but it's not a substitute for a professional diagnosis. It's always advisable to consult with a mental health pro-fessional if you suspect that you or someone you know may be struggling with depression.

Enhancing Li+ recovery in brine mining: integrating next-gen emotional AI and explainable ML to predict adsorption energy in crown ether-based hierarchical nanomaterials.

Artificial intelligence (AI) is being employed in brine mining to enhance the extraction of lithium, vital for the manufacturing of lithium-ion batteries, through improved recovery efficiencies and the reduction of energy consumption. An innovative approach was proposed combining Emotional Neural Networks (ENN) and Random Forest (RF) algorithms to elucidate the adsorption energy (AE) (kcal mol-1) of Li+ ions by utilizing crown ether (CE)-incorporated honeycomb 2D nanomaterials. The screening and feature engineering analysis of honeycomb-patterned 2D materials and individual CE were conducted through Density Functional Theory (DFT) and Gaussian 16 simulations. The selected honeycomb-patterned 2D materials encompass graphene, silicene, and hexagonal boron nitride, while the specific CEs evaluated are 15-crown-5 and 18-crown-6. The crown-passivated 2D surfaces held a significant adsorption site through van der Waals forces for efficient recovery of Li+ ions. ENN predicted the targeted adsorption sites with high precision and minimal deviation. The eTAI (XAI) based Shapley Additive exPlanations (SHAP) was also explored for insight into the feature importance of CE embedded 2D nanomaterials for the recovery of Li+ ions. The extreme gradient boosting algorithm (XGBoost) model demonstrated a RT-2-MAPE = 0.4618% and ENN-2-MAPE = 0.4839% for the feature engineering analysis. This research would be an insight into the AI-driven nanotechnology that presents a viable and sustainable approach for the extraction of natural resources through the application of brine mining.

The Anchoring Effect, Algorithmic Fairness, and the Limits of Information Transparency for Emotion Artificial Intelligence

Emotion artificial intelligence (AI) is shown to vary systematically in its ability to accurately identify emotions, and this variation creates potential biases. In this paper, we conduct an experiment involving three commercially available emotion AI systems and a group of human labelers tasked with identifying emotions from two image data sets. The study focuses on the alignment between facial expressions and the emotion labels assigned by both the AI and humans. Importantly, human labelers are given the AI’s scores and informed about its algorithmic fairness measures. This paper presents several key findings. First, the labelers’ scores are affected by the emotion AI scores, consistent with the anchoring effect. Second, information transparency about the AI’s fairness does not uniformly affect human labeling across different emotions. Moreover, information transparency can even increase human inconsistencies. Plus, significant inconsistencies in the scoring among different emotion AI models cast doubt on their reliability. Overall, the study highlights the limitations of individual decision making and information transparency regarding algorithmic fairness measures in addressing algorithmic fairness. These findings underscore the complexity of integrating emotion AI into practice and emphasize the need for careful policies on emotion AI.

Voice Cloning Using Artificial Intelligence and Machine Learning: A Review

This paper represents a thorough method for integrating emotions, texttospeech conversion, and state of the art voice cloning. The paper focuses on novel background noise adaptation, emotional voice synthesis, and multi-speaker voice cloning for better speech synthesis. The synthesis of emotive voices, multi-speaker voice cloning, and creative methods for modifying background noise to improve speech synthesis quality are among the topics covered in this study. Additionally, the study explores the domain of emotional artificial intelligence by adding a variety of emotions to artificial voices, improving user engagement through sympathetic reactions. The study also looks at how background noise can be altered to change it from a disturbing to a silent, non-disruptive state. The texttospeech systems usability in noisy conditions is greatly enhanced by this improvement. By integrating these components, the project makes a substantial contribution to text to speech, emotional AI, and voice cloning, creating new avenues for human-computer connection.

Розпізнавання емоцій користувача із використанням штучного інтелекту

The article explains the face recognition process. The stages of recognition of emotions on the face and groups according to the field of application of FER are described. The main types of emotional artificial intelligence are distinguished. Seven recognized universal emotions are considered. A custom facial recognition script has been developed. Realized the functions of reading and detecting emotions live through a webcam using DeepFace.The research examines the development of the program "Recognition of user emotions using artificial intelligence." The following tasks were implemented in the work: writing the function of detecting a person in a photo or video and teaching the program to distinguish different facial emotions: happiness, sadness, anger, etc., using the DeepFace and OpenCV face recognition package. The developed computer program can be used in various aspects of everyday life: provision of personalized services, health care, employment, education, public safety. The program was tested by students of higher education major 122 "Computer science" of the Central Ukrainian State University named after V. Vinnichenko.

On the Notion of Emotional Truth

In this paper, we propose an explication of the notion of emotional (pragmatic) truth due to defining performative propositions on non-Archimedean models [Formula: see text]. In these models, it is possible to fix an appropriate behavioral scenario for each atomic performative proposition. Meanwhile, each performative proposition [Formula: see text] has an appropriate form of cognitive bias [Formula: see text] changing the meaning of [Formula: see text], but this bias is the result or foundation for interaction with the outside world. Therefore, instead of the usual domain [Formula: see text] of realization of atomic informative propositions [Formula: see text], which can be finite, we always have an infinite domain [Formula: see text] of realization of performative propositions [Formula: see text], where [Formula: see text] is an informative proposition, distorted by some kind of cognitive biases [Formula: see text]. This semantics allows us to shift the focus from artificial intelligence (AI) to emotional artificial intelligence (EAI).

Affective computing technology for fostering an emotionally healthy workplace

Purpose This paper aims to present affective computing or Emotion AI in the context of work and how organizational leaders such as managers and human resource (HR) professionals can implement this technology to foster an emotionally healthy workplace. Design/methodology/approach The authors provide a current overview of affective computing technology through definitions, examples and general use cases. This is in light of the current scrutiny on artificial intelligence (AI) use broadly across society. The authors address this from a research perspective and show how this advanced AI tool can be implemented in organizations for the benefit of employees. Findings Affective computing or Emotion AI is still relatively unknown, and yet, it is already part of our daily lives. Emotion AI platforms have the potential to be an essential part of HR tools. It is crucial, however, to use this technology in an ethical and responsible manner. Originality/value There is little awareness and understanding of use cases of affective computing tools in organizations, particularly for the well-being of the workforce. This paper provides HR leaders, managers and researchers with an overview of the origins of the field and major considerations for responsibly implementing Emotion AI to support employee mental health.

Emotional Artificial Intelligence as a Tool for Human-Machine Interaction

Introduction. One of the trends of global importance is the artificial intelligence (AI) and its innovations. One of such innovations has become emotional artificial intelligence (emotional AI/AI), it is called a revolutionary technology that can identify human emotions, process them in a timely manner and react “properly”. Experts consider emotional AI as an instrument that provides emotionally-oriented human–machine communication. The article discusses the specifics of emotional AI, achievements, potential opportunities, development prospects.Methodology and sources. The methodology of philosophical, socio-psychological, comparative and interdisciplinary approaches is used. The sources used in the article are: special literature of domestic and foreign authors (B. Goertzel, D. Goleman, R. Picard,D.I. Dubrovsky, E.M. Proidakov) scientific research, publications and websites devoted to emotional artificial intelligence, and its features (Aliya Green Emotional artificial intelligence: changing the human world for the better).Results and discussion. The relevance of the topic of emotional AI determined the need to refer to the concept of ”emotional intelligence” (EI) as the basic basis of emotional artificial intelligence, which allowed us to show the essential characteristics of human emotional intelligence, its difference from AI. Emotional artificial intelligence is an innovation of modern AI, its main actors are anthropomorphic robots, text, voice chatbots and video bots, which are already actively demonstrating to the public their knowledge and skills in the field of psychology of emotions, which are being improved within the framework of the current AI.Conclusion. Currently, there is a gradual process of teaching emotional AI to interact with a person, and although these achievements are not great yet, EII is gradually developing in accordance with the challenges of new realities within the specifics of modern applied AI. However, in the digital age, human–machine and machine-to-human communication is an interconnected process that should be aimed at building both utilitarian–useful and partnership relations in the practices of their interaction, which meets the requirements of the era and leads to further progress of AI – to the creation of a new, common AI – “human AI the level” which is supposed to greatly expand the capabilities of a person and society as a whole.

Data Subjects' Perspectives on Emotion Artificial Intelligence Use in the Workplace: A Relational Ethics Lens

The workplace has experienced extensive digital transformation, in part due to artificial intelligence's commercial availability. Though still an emerging technology, emotional artificial intelligence (EAI) is increasingly incorporated into enterprise systems to augment and automate organizational decisions and to monitor and manage workers. EAI use is often celebrated for its potential to improve workers' wellbeing and performance as well as address organizational problems such as bias and safety. Workers subject to EAI in the workplace are data subjects whose data make EAI possible and who are most impacted by it. However, we lack empirical knowledge about data subjects' perspectives on EAI, including in the workplace. To this end, using a relational ethics lens, we qualitatively analyzed 395 U.S. adults' open-ended survey (partly representative) responses regarding the perceived benefits and risks they associate with being subjected to EAI in the workplace. While participants acknowledged potential benefits of being subject to EAI (e.g., employers using EAI to aid their wellbeing, enhance their work environment, reduce bias), a myriad of potential risks overshadowed perceptions of potential benefits. Participants expressed concerns regarding the potential for EAI use to harm their wellbeing, work environment and employment status, and create and amplify bias and stigma against them, especially the most marginalized (e.g., along dimensions of race, gender, mental health status, disability). Distrustful of EAI and its potential risks, participants anticipated conforming to (e.g., partaking in emotional labor) or refusing (e.g., quitting a job) EAI implementation in practice. We argue that EAI may magnify, rather than alleviate, existing challenges data subjects face in the workplace and suggest that some EAI-inflicted harms would persist even if concerns of EAI's accuracy and bias are addressed.

Values in Emotion Artificial Intelligence Hiring Services: Technosolutions to Organizational Problems

Despite debates about emotion artificial intelligence's (EAI) validity, legality, and social consequences, EAI is increasingly present in the high stakes context of hiring, with potential to shape the future of work and the workforce. The values laden in technology play a significant role in its societal impact.We conducted qualitative content analysis on the public-facing websites (N=229) of EAI hiring services. We identify the organizational problems that EAI hiring services claim to solve and reveal the values emerging in desired EAI uses as promoted by EAI hiring services to solve organizational problems. Our findings show that EAI hiring services market their technologies as technosolutions to three purported organizational hiring problems: 1) hiring (in)accuracy, 2) hiring (mis)fit, and 3) hiring (in)authenticity. We unpack these problems to expose how these desired uses of EAI are legitimized by the corporate ideals of data-driven decision making, continuous improvement, precision, loyalty, and stability. We identify the unfair and deceptive mechanisms by which EAI hiring services claim to solve the purported organizational hiring problems, suggesting that they unfairly exclude and exploit job candidates through EAI's creation, extraction, and affective commodification of a candidate's affective value through pseudoscientific approaches. Lastly, we interrogate EAI hiring service claims to reveal the core values that underpin their stated desired use: techno-omnipresence, techno-omnipotence, and techno-omniscience. We show how EAI hiring services position desired use of their technology as a moral imperative for hiring organizations with supreme capabilities to solve organizational hiring problems, then discuss implications for fairness, ethics, and policy in EAI-enabled hiring within the US policy landscape.

SENTIMENT ANALYSIS OF STUDENTS’ FEEDBACK ON INSTITUTIONAL FACILITIES USING TEXT-BASED CLASSIFICATION AND NATURAL LANGUAGE PROCESSING (NLP)

In this study, sentiment analysis, commonly referred to as opinion mining or emotional artificial intelligence (AI), is used to obtain student comments about institutional facilities. To ascertain whether an online text has a good, negative, or neutral emotional tone, it must first be analyzed. Natural Language Processing (NLP) includes the subfield of sentiment analysis, and NLP can be used to categorize and extract information with the aid of machine learning methods. Finding out if students are content with the amenities or services provided is important since, in an educational setting, they are consumers. This study of the resources and services that the libraries offer evaluated college students' perceptions of the books, audio CDs, and video CDs, the services provided by the library staff, and the personal computers that are made available to them as part of their facilities. In the current study, surveys were carried out to gather information and assess how well the needs of the students are addressed. On Kaggle.com, secondary data from a North Indian institution was used in the experiments. The research approach was sentiment analysis using a machine learning framework. The F1-score, a harmonic mean of precision and recall based on the attitudes evaluated by the algorithms, was calculated using a text-based classification method using Naive Bayes Multinomial and Support Vector machine learning algorithms. The findings offer views on how successfully library facilities are managed in the form of student sentiments.

Unifying Algorithmic and Theoretical Perspectives: Emotions in Online Reviews and Sales

Emotion artificial intelligence, the algorithm that recognizes and interprets various human emotions beyond valence (positive and negative polarity), is still in its infancy but has attracted attention from industry and academia. Based on discrete emotion theory and statistical language modeling, this work proposes an algorithm to enable automatic domain-adaptive emotion lexicon construction and multidimensional emotion detection in texts. Using a large-scale dataset of China’s movie market from 2012 to 2018, we constructed and validated a domain-specific emotion lexicon and demonstrated the predictive power of eight discrete emotions (i.e., surprise, joy, anticipation, love, anxiety, sadness, anger, and disgust) in online reviews on box office sales. We found that representing overall emotions through discrete emotions yields higher prediction accuracy than valence or latent emotion variables generated by topic modeling. To understand the source of the predictive power from a theoretical perspective and to test the cross-culture generalizability of our prediction study, we further conducted an experiment in the U.S. movie market based on theories on emotion, judgment, and decision-making. We found that discrete emotions, mediated by perceived processing fluency, significantly affect the perceived review helpfulness, which further influences purchase intention. Our work shows the economic value of emotions in online reviews, generates insight into the mechanism of their effects, and has managerial implications for online review platform design, movie marketing, and cinema operations.

The Fairness Principle: A Tool to Protect Childrens Rights in Their Interaction with Emotional AI in Educational Settings

Emotional artificial intelligence (EAI) is increasingly being used in educational settings. This raises several contentious issues regarding children’s rights and freedoms. This article evaluates the impact of EAI on children’s rights, focusing on freedom of thought (FoT), non-discrimination, privacy, and data protection. These rights are evaluated in turn before focusing on the fairness principle in Article 5(1)(a) of the General Data Protection Regulation (GDPR), an ‘underutilized’ principle that lies at the heart of the United Kingdom (UK) Age-Appropriate Design Code’s standard: best interests of the child. We argue that the use of EAI in education contexts contradicts the fairness principle and that EAI represents a potentially serious threat to children’s rights to FoT, non-discrimination, privacy, and data protection and is an issue which requires urgent attention from policymakers and regulators in the UK. Fairness, Emotional Artificial Intelligence, GDPR, Privacy, Freedom of Thought, Education

An Efficient Machine Learning-Based Emotional Valence Recognition Approach Towards Wearable EEG.

Emotion artificial intelligence (AI) is being increasingly adopted in several industries such as healthcare and education. Facial expressions and tone of speech have been previously considered for emotion recognition, yet they have the drawback of being easily manipulated by subjects to mask their true emotions. Electroencephalography (EEG) has emerged as a reliable and cost-effective method to detect true human emotions. Recently, huge research effort has been put to develop efficient wearable EEG devices to be used by consumers in out of the lab scenarios. In this work, a subject-dependent emotional valence recognition method is implemented that is intended for utilization in emotion AI applications. Time and frequency features were computed from a single time series derived from the Fp1 and Fp2 channels. Several analyses were performed on the strongest valence emotions to determine the most relevant features, frequency bands, and EEG timeslots using the benchmark DEAP dataset. Binary classification experiments resulted in an accuracy of 97.42% using the alpha band, by that outperforming several approaches from literature by ~3-22%. Multiclass classification gave an accuracy of 95.0%. Feature computation and classification required less than 0.1 s. The proposed method thus has the advantage of reduced computational complexity as, unlike most methods in the literature, only two EEG channels were considered. In addition, minimal features concluded from the thorough analyses conducted in this study were used to achieve state-of-the-art performance. The implemented EEG emotion recognition method thus has the merits of being reliable and easily reproducible, making it well-suited for wearable EEG devices.

Can Emotion AI Keep You Healthier?

Even for humans, it can be challenging to recognize, interpret, and respond to emotions. Can artificial intelligence (AI) do any better? Technologies often referred to as "emotion AI" detect and analyze facial expressions, voice patterns, muscle activity, and other behavioral and physiological signals associated with emotions.

How do you solve a problem like Alexa?

Emotional Artificial Intelligence (EAI) is emerging as a mainstream technology. With the increasing use of EAI in different sectors, novel legal and ethical concerns are being raised. This paper focuses on an entertainment and creative content delivery recommendation system application, namely Amazon Alexa’s use of Neural-Text-to-Speech (NTTS) technology to capture and respond to users’ perceived emotions based on their voice. It also enables Alexa to play music based on, among other elements, its perception of users’ emotions. Given the recognised impact ‘music‘ has on emotions and Alexa‘s increasing involvement with big music sector actors, we are particularly concerned about how this domain can enable the manipulation of users. We use this example to highlight problems with the approach to AI regulation taken in the proposed EU Artificial Intelligence Act (AIA).

Text Sentiment Analysis Based on Multi-Layer Bi-Directional LSTM with a Trapezoidal Structure

Sentiment analysis, commonly called opinion mining or emotion artificial intelligence (AI), employs biometrics, computational linguistics, natural language processing, and text analysis to systematically identify, extract, measure, and investigate affective states and subjective data. Sentiment analysis algorithms include emotion lexicon, traditional machine learning, and deep learning. In the text sentiment analysis algorithm based on a neural network, multi-layer Bi-directional long short-term memory (LSTM) is widely used, but the parameter amount of this model is too huge. Hence, this paper proposes a Bi-directional LSTM with a trapezoidal structure model. The design of the trapezoidal structure is derived from classic neural networks, such as LeNet-5 and AlexNet. These classic models have trapezoidal-like structures, and these structures have achieved success in the field of deep learning. There are two benefits to using the Bi-directional LSTM with a trapezoidal structure. One is that compared with the single-layer configuration, using the of the multi-layer structure can better extract the high-dimensional features of the text. Another is that using the trapezoidal structure can reduce the model’s parameters. This paper introduces the Bi-directional LSTM with a trapezoidal structure model in detail and uses Stanford sentiment treebank 2 (STS-2) for experiments. It can be seen from the experimental results that the trapezoidal structure model and the normal structure model have similar performances. However, the trapezoidal structure model parameters are 35.75% less than the normal structure model.