Emotional Model Research Articles

Bimodal deep learning model for subjectively enhanced emotion classification in films

This research delves into the concept of color grading in film, focusing on how color influences the emotional response of the audience. The study commenced by recalling state-of-the-art works that process audio–video signals and associated emotions by machine learning. Then, assumptions of subjective tests for refining and validating an emotion model for assigning specific emotional labels to selected film excerpts were presented. The insights gained from these subjective evaluations facilitated the creation of a comprehensive database of movie excerpts. This database was subsequently employed to both train and evaluate the efficacy of deep learning models. The latter half of the study shifts focus to the intelligent analysis of audio and video signals that form film excerpts. This involved exploring diverse methodologies for parameterizing these signals. Models that demonstrated the highest accuracy on the test dataset on audio/video only were amalgamated to forge a bimodal model, which integrates both audio and video signals for emotion classification. The bimodal model exhibited superior accuracy in tests compared to a model that solely relied on video signal classification. This enhancement in performance was achieved with only a marginal increase in the complexity and the number of parameters within the model.

Analysis of patient health questionnaire-9 (PHQ-9) based depression prevalence according to a discordance between quantitative urinary cotinine levels and self-report of second-hand smoke exposure among adults: A cross-sectional study

BackgroundSecond-hand smoke (SHS) exposure appears to be more common among individuals with depression. However, self-report of SHS exposure is an inaccurate classification compared to confirming SHS exposure using urinary cotinine (UC). Additionally, the dose-response relationship between depression and UC is controversial. MethodsThe severe stress rate and depression prevalence was estimated among 14530 Korean participants aged ≥19 years using data patient health questionnaire-9 (PHQ-9) and on UC from the Korean National Health and Nutrition Examination Survey. Measured UCs were divided into four categories: UC– (≤0.3 μg/L), UC± (0.4 μg/L–0.9 μg/L), UC+ (1.0 μg/L–11.9 μg/L), and UC++ (≥12.0 μg/L). ResultsAbout 55.0 % participants were female and participants’ mean age was 51.1 years. Non-smokers were 80.3 %. Among non-smokers, non-SHS exposure participants (SR–) and SHS exposure participants (SR+) were 83.0 % and 17.0 %, respectively. When UC– was used as the reference subgroup, the UC++ subgroup showed a higher depression prevalence, whereas the UC ± subgroup showed a lower prevalence. In the same UC categories, the depression prevalence and severe stress rate were higher among females than among males. Furthermore, the SR + subgroup had a higher severe stress rate than the SR– subgroup. ConclusionsOur study showed a paradoxical reduction in the depression prevalence and severe stress rate in the UC ± subgroup compared to the UC– subgroup. Additionally, the dose-response relationship between the SHS exposure biomarker and the depression prevalence was not linear. Our study indicates that an emotional stress-based model may be more appropriate for explaining the relationship between depression and SHS exposure.

A Visual Approach to Tracking Emotional Sentiment Dynamics in Social Network Commentaries

The expansion of social media has unlocked a real-time barometer of public opinion. This paper introduces a novel framework to analyze sentiment shifts in social network comment sections, a reflection of the broader public discourse over time. Leveraging a pre-trained uncased RoBERTa model, we predict emotional scores from user comments, mapping these to key sentiment trends such as Approval, Toxicity, Obscenity, Threat, Hate, Offensive, and Neutral. Our methodology employs machine learning techniques to train a dataset that connects emotional scores with these trends, generating trend probability scores. We utilize a bottom-up recursive algorithm to aggregate emotional scores within comment threads, enabling the prediction of trend scores using three distinct aggregation methods. The results demonstrate that our emotional prediction model achieves an AUC of 0.92, and XGBoost stands out with an F1 score exceeding 0.40. Our research elucidates the temporal evolution of online public sentiment, enhancing the understanding of digital social dynamics and offering insights for strategic online interaction, intervention, and content moderation.

FACIAL IMAGE CAPTIONING USING DNN

Facial analysis, encompassing emotion, age, and gender detection, shows potential in various applications such as human-computer interaction, business, security, and health. This study delves into the development and evaluation of a deep neural network (DNN) model for facial emotion, age, and gender detection. Utilizing a convolutional neural network (CNN) architecture trained on diverse datasets for each task, our model proves effective in predicting facial features. The accuracy of needs assessment is X%, the marginal error (MAE) of age estimation is Y years, and the accuracy of gender classification is Z%.

The emotional boundary seesaw model to help cancer clinical nurse specialists work with patients who challenge them.

Receiving a diagnosis of cancer and receiving treatment can be physically and psychologically onerous for cancer patients. Alongside their nursing skills, cancer clinical nurse specialists (CCNSs) are trained to provide psychological support, assessment and intervention following a cancer diagnosis. Mental health specialists, usually clinical psychologists, provide clinical supervision to CCNSs to support them with this. One of the regular themes that comes up in clinical supervision with CCNSs is how to maintain a supportive and therapeutic relationship with 'challenging' patients. These patients may be considered challenging because they are inconsistent, demanding or ambivalent about their care, which puts a strain on the relationship with their CCNS. By using the emotional seesaw model to reflect on and attend to the emotional boundaries of the relationship, within clinical supervision, these challenging relationships can be understood and managed, and better outcomes for the patient can be achieved.

Agent-based Emotional Persuasion in the Psychological Counselling System for University Students

Chinese college students have a high demand for psychological counselling, and a survey by the National Institute of Psychology shows that college students are resistant to face-to-face psychological counselling. This paper integrates the emotion model with the emotion fading factor and emotion evaluation factor based on Agent and builds the models to give Agent the complex emotional characteristics of humans. This paper applies the agent-based emotional persuasion to the psychological counselling system and uses the emotional persuasion model to give full play to the influence of the Agent s decision-making process in the psychological counselling system.

Hybrid Natural Language Processing Model for Sentiment Analysis during Natural Crisis

This paper introduces a novel natural language processing (NLP) model as an original approach to sentiment analysis, with a focus on understanding emotional responses during major disasters or conflicts. The model was created specifically for Croatian and is based on unigrams, but it can be used with any language that supports the n-gram model and expanded to multiple word sequences. The presented model generates a sentiment score aligned with discrete and dimensional emotion models, reliability metrics, and individual word scores using affective datasets Extended ANEW and NRC WordEmotion Association Lexicon. The sentiment analysis model incorporates different methodologies, including lexicon-based, machine learning, and hybrid approaches. The process of preprocessing includes translation, lemmatization, and data refinement, utilized automated translation services as well as the CLARIN Knowledge Centre for South Slavic languages (CLASSLA) library, with a particular emphasis on diacritical mark correction and tokenization. The presented model was experimentally evaluated on three simultaneous major natural crises that recently affected Croatia. The study’s findings reveal a significant shift in emotional dimensions during the COVID-19 pandemic, particularly a decrease in valence, arousal, and dominance, which corresponded with the two-month recovery period. Furthermore, the 2020 Croatian earthquakes elicited a wide range of negative discrete emotions, including anger, fear, and sadness, with the recuperation period much longer than in the case of COVID-19. This study represents an advancement in sentiment analysis, particularly in linguistically specific contexts, and provides insights into the emotional landscape shaped by major societal events.

FER-BHARAT: a lightweight deep learning network for efficient unimodal facial emotion recognition in Indian context

Humans' ability to manage their emotions has a big impact on their ability to plan and make decisions. In order to better understand people and improve human–machine interaction, researchers in affective computing and artificial intelligence are investigating the detection and recognition of emotions. However, different cultures have distinct ways of expressing emotions, and the existing emotion recognition datasets and models may not effectively capture the nuances of the Indian population. To address this gap, this study proposes custom-built lightweight Convolutional Neural Network (CNN) models that are optimized for accuracy and computational efficiency. These models are trained and evaluated on two Indian emotion datasets: The Indian Spontaneous Expression Dataset (ISED) and the Indian Semi Acted Facial Expression Database (iSAFE). The proposed CNN model with manual feature extraction provides remarkable accuracy improvement of 11.14% for ISED and 4.72% for iSAFE datasets as compared to baseline, while reducing the training time. The proposed model also surpasses the accuracy produced by pre-trained ResNet-50 model by 0.27% ISED and by 0.24% for the iSAFE dataset with significant improvement in training time of approximately 320 s for ISED and 60 s for iSAFE dataset. The suggested lightweight CNN model with manual feature extraction offers the advantage of being computationally efficient and more accurate compared to pre-trained model making it a more practical and efficient solution for emotion recognition among Indians.

Detecting Users' Emotional States during Passive Social Media Use

The widespread use of social media significantly impacts users' emotions. Negative emotions, in particular, are frequently produced, which can drastically affect mental health. Recognizing these emotional states is essential for implementing effective warning systems for social networks. However, detecting emotions during passive social media use---the predominant mode of engagement---is challenging. We introduce the first predictive model that estimates user emotions during passive social media consumption alone. We conducted a study with 29 participants who interacted with a controlled social media feed. Our apparatus captured participants' behavior and their physiological signals while they browsed the feed and filled out self-reports from two validated emotion models. Using this data for supervised training, our emotion classifier robustly detected up to 8 emotional states and achieved 83% peak accuracy to classify affect. Our analysis shows that behavioral features were sufficient to robustly recognize participants' emotions. It further highlights that within 8 seconds following a change in media content, objective features reveal a participant's new emotional state. We show that grounding labels in a componential emotion model outperforms dimensional models in higher-resolutional state detection. Our findings also demonstrate that using emotional properties of images, predicted by a deep learning model, further improves emotion recognition.

Influence of Severity and Susceptibility Information Content on Users’ Psychological Response in Emergencies—The Moderating Effect of Regulatory Focus

Information content is crucial for public opinion governance, yet its psychological mechanisms remain unclear. This paper endeavors to bridge this knowledge gap by examining the influence of information content on users’ psychological responses. Drawing insights from the heuristic-systematic model (HSM) and regulatory focus theory, we constructed a novel cognitive and emotional response model. Regression analysis reveals that susceptibility strongly influences primary cognitive responses, while severity affects secondary ones. Susceptibility has a stronger impact on emotional responses than severity. Promotion-focus users are less influenced by severity and susceptibility than prevention-focus users. These results offer valuable insights for emergency management strategies and enhance individuals’ resilience in coping with the psychological toll of emergencies.

How can entrepreneurial obstacles improve entrepreneurs’ psychological well-being? The role of emotional ambivalence

ABSTRACT Entrepreneurial obstacles present both threats to survival and opportunities for learning. By integrating the double-edged nature of obstacles into the evaluative space model of emotion, this study examines the effects of entrepreneurial obstacles on entrepreneurs’ emotional ambivalence and their psychological well-being. Through survey among 324 new ventures in China, we find that although entrepreneurial obstacle is directly detrimental to psychological well-being, it can also trigger entrepreneurs’ emotional ambivalence which has a U-shaped relationship with psychological well-being. Moreover, demand uncertainty increases entrepreneurs’ ambivalent emotional response to entrepreneurial obstacles. This study identifies a self-healing path between entrepreneurial obstacles and entrepreneurs’ psychological well-being.

Do Societies Have Emotions?

In recent decades, there has been a growing interest in the study of emotions, demonstrating their importance in understanding diverse human and social dynamics. However, as this field of study develops and diversifies, a structural yet simple question remains virtually undiscussed: Is it really possible to say that societies possess emotions or a certain sensitivity akin to individual experiences? This exploratory study, based on documentary analysis, endeavours to identify and examine emotional patterns across six different periods, spanning from the Middle Ages to the present day. Among the eight secondary findings, three fundamental conclusions stand out: (1) throughout history, societies have experienced different emotional atmospheres, sometimes simultaneously; (2) although societies generally propose an emotional model to follow, the reality does not always conform to it; and (3) it is mainly through the culture that society creates a certain emotional harmony, allowing the social body to remain cohesive and develop, thus postponing or preventing its disintegration. This study aims to offer a modest contribution to the complex and under-explored discussion on the correlation between specific emotional climates and particular social contexts.

Role of self-esteem and personal mastery on the association between social support and resilience among COVID-19 long haulers

Although the COVID-19 mortality rate is declining, the number of individuals dealing with persistent COVID-19 symptoms is increasing worldwide, making long COVID a global public health concern. People with long COVID (long haulers) often deal with physical and mental stressors. Long haulers’ psychological resilience could play a key role in coping with these stressors in intercorrelation with psychosocial resources. The current study aims to test a hypothesized relationship between social support and its functions (i.e., instrumental and emotional) and the resilience of long haulers through serial mediation by personal mastery and self-esteem. A cross-sectional and self-administered online survey was conducted among 460 individuals with long COVID recruited from COVID-19 Facebook support groups in the United States. Analyzing data indicated a positive correlation between social support and the resilience of long haulers. Structural equation modeling suggested that self-esteem and personal mastery fully mediated the association between social (instrumental) support and resilience. Personal mastery also mediated the association between self-esteem and resilience in social (instrumental) support models. However, in the emotional support model, the indirect effect was non-significant for the mediation by personal mastery and self-esteem. Findings suggest that social support, mainly instrumental support, may protect long haulers by promoting their resilience through self-esteem and personal mastery. This study emphasizes the importance of including social support services in designing programs for COVID-19 long haulers.

Enhanced Sarcasm and Emotion Detection Through Unified Model of Transformer and FCNets

In this study, we present a unified model for efficient emotion and sarcasm identification in many languages. We accomplish resilient performance by integrating hybrid transformers and fully connected networks (FCNets) in our methodology. Extracting datasets in several languages, doing exploratory data analysis, preprocessing, extracting features, training and testing models, and finally deploying the system are all essential processes in the suggested strategy. Using a variety of models throughout training, we apply deep learning approaches to the problem of sarcasm and emotion recognition. The Flask framework chooses the model with the best performance to deploy. When applied to situations involving many languages, our method shows to be far more efficient and effective at identifying instances of sarcasm and emotion. Our model gives accuracy of 93.130 % for identifying sarcasm and emotion.

Early Church as Utopian Cult: The Case of Hebrews

Abstract This article uses cultic studies to explicate the development of the earliest churches, using Hebrews as a case study. “Cult” is defined as a group that suppresses individuality and critical thinking, demands intense commitment and loyalty to an individual or ideology, and curtails or eliminates personal autonomy to be replaced by the cult’s worldview. The cult-like attributes of the early ecclesia are summarized, followed by a brief introduction to cultic studies, with particular reference to the characteristics of contemporary cults, suggesting that an under-appreciated aspect of cultic appeal is the offer of utopian promises. Finally, the bite (Behavior, Information, Thoughts, Emotions) model of coercive control is applied to the hortatory sections of Hebrews and the utopian hopes held out by the author. The article concludes that the community behind Hebrews shows many characteristics of a high demand cult as defined by cultic studies.

Research on emotion modeling of intelligent agents in earthquake evacuation simulation

In recent years, virtual reality (VR) has been widely used in emergency drills, skills training and other fields of science education. However, existing VR emergency training platforms lack intelligent agents endowed with emotional and cognitive abilities, making them challenging to evoke user emotions and achieve immersion. Moreover, existing emotion contagion methods of virtual agents lack analysis on whether the emotions of virtual agents can infect users in the real world. Therefore, we proposed an emotional cognitive model (ECM) that simulates the emotional contagion of intelligent agents in a VR earthquake emergency training platform. To evaluate the proposed model, we conducted a user study requiring the user to control an avatar for evacuation training during a simulated earthquake. The user’s brain signals are measured using EEG, fNIRS, and eye-tracking devices to analyze whether the user was affected by the emotions of other intelligent agents. The results show that intelligent agents with emotional cognition can evoke user’s emotions in earthquake emergency training.

A historical approach to models of emotional laterality

In this paper I discuss the main models that have tried to explain brain asymmetries for emotions. The first models, based on clinical observations, proposed either a general right hemisphere dominance for emotions (the’right hemisphere’) model or a different specialization of the right hemisphere for negative and of the left hemisphere for positive emotions (the‘valence’ model). In more recent times new models, based on partly modified versions of the previous ones have been proposed. The revised version of the ‘valence’ model, labeled the ‘approach-avoidance’ model maintained that hemispheric asymmetries are not related to the valence of the emotional stimulus but to the motivational (approach vs avoidance) system that is engaged by that stimulus. On the contrary, revised versions of the ‘right hemisphere’ hypothesis proposed graded versions of this model, maintaining that only some kinds or some levels of emotions are clearly right lateralized. One version of these models (the’emotion type hypothesis’) assumed that only elementary basic emotions should be subsumed by the right hemisphere, wheres more complex social emotions should be subtended by the left hemisphere. The other version (the ‘schematic level of emotion hypothesis’) assumed that the right hemisphere should subsume only the basic ‘schematic’ level of emotions, characterized by an automatic and unconscious processing, whereas the more propositional and conscious ‘conceptual’ level could be less lateralized or subsumed by the left hemisphere. This last model is supported by the obsevation that the right hemisphere reveals a modus operandi (i.e. a prevalence of the ‘automatic’ over the ‘intentional’ and of the ‘unconscious’ over the ‘conscious’ functional processing) that is typical of the ‘schematic level of emotions.

Non-speech emotion recognition based on back propagation feed forward networks

Non-speech emotion recognition involves identifying emotions conveyed through non-verbal vocalizations such as laughter, crying, and other sound signals, which play a crucial role in emotional expression and transmission. This paper employs a nine-category discrete emotion model encompassing happy, sad, angry, peaceful, fearful, loving, hateful, brave, and neutral. A proprietary non-speech dataset comprising 2337 instances was utilized, with 384-dimensional feature vectors extracted. The traditional Backpropagation Neural Network (BPNN) algorithm achieved a recognition rate of 87.7% on the non-speech dataset. In contrast, the proposed Whale Optimization Algorithm - Backpropagation Neural Network (WOA-BPNN) algorithm, applied to a self-made non-speech dataset, demonstrated a remarkable accuracy of 98.6%. Notably, even without facial emotional cues, non-speech sounds effectively convey dynamic information, and the proposed algorithm excels in their recognition. The study underscores the importance of non-speech emotional signals in communication, especially with the continuous advancement of artificial intelligence technology. The abstract thus encapsulates the paper’s focus on leveraging AI algorithms for high-precision non-speech emotion recognition.

Immersive Emotion Analysis in VR Environments: A Sensor-Based Approach to Prevent Distortion

As virtual reality (VR) technology advances, research has focused on enhancing VR content for a more realistic user experience. Traditional emotion analysis relies on surveys, but they suffer from delayed responses and decreased immersion, leading to distorted results. To overcome these limitations, we propose an emotion analysis method using sensor data in the VR environment. Our approach can take advantage of the user’s immediate response and not reduce immersion. Linear regression, classification analysis, and tree-based methods were applied to electrocardiogram and galvanic skin response (GSR) sensor data to measure valence and arousal values. We introduced a novel emotional dimension model by analyzing correlations between emotions and the valence and arousal values. Experimental results demonstrated the highest accuracy of 77% and 92.3% for valence and arousal prediction, respectively, using GSR sensor data. Furthermore, an accuracy of 80.25% was achieved in predicting valence and arousal using nine emotions. Our proposed model improves VR content through more accurate emotion analysis in a VR environment, which can be useful for targeting customers in various industries, such as marketing, gaming, education, and healthcare.

Liking music with and without sadness: Testing the direct effect hypothesis of pleasurable negative emotion.

Negative emotion evoked in listeners of music can produce intense pleasure, but we do not fully understand why. The present study addressed the question by asking participants (n = 50) to self-select a piece of sadness-evoking music that was loved. The key part of the study asked participants to imagine that the felt sadness could be removed. Overall participants reported performing the task successfully. They also indicated that the removal of the sadness reduced their liking of the music, and 82% of participants reported that the evoked sadness also adds to the enjoyment of the music. The study provided evidence for a "Direct effect hypothesis", which draws on the multicomponent model of emotion, where a component of the negative emotion is experienced as positive during music (and other aesthetic) experiences. Earlier evidence of a mediator, such as 'being moved', as the source of enjoyment was reinterpreted in light of the new findings. Instead, the present study applied a semantic overlap explanation, arguing that sadness primes emotions that share meaning with sadness, such as being-moved. The priming occurs if the overlap in meaning is sufficient. The degree of semantic overlap was defined empirically. The present study therefore suggests that mediator-based explanations need to be treated with caution both as a finding of the study, and because of analytic limitations in earlier research that are discussed in the paper.