Multiple Emotions Research Articles

Capturing multiple emotions from conversational data using fine tuned transformers

Emotion detection is one of the crucial topics of Natural Language Processing (NLP) in recent years, and now one of the biggest motivating factors in correct identification and interpretation of a wide range of emotional expressions in textual data. This research examines the use of Bidirectional Encoder Representations from Transformers (BERT), a trained transformer model for emotion detection in textual data. This analysis evaluates how good BERT is at identifying emotions such as surprise, anger, fear, happiness and sadness compared with ordinary machine learning as well as deep learning techniques. The addition of weighted emotions approach enhances the model performance and gives a deeper emotional context awareness making it more effective to deal with complicated emotional utterances. The purpose of the research is to develop a fine-tuned BERT model with a weighted emotion framework to enhance the accuracy of emotion classification in conversational text. The context is improving emotion recognition in scenarios where multiple emotions co-exist, addressing limitations in traditional models by capturing subtle and overlapping emotional expressions. In terms of training the methods multiple datasets are considered and also the research examines various models performance. The article further discusses possible application areas of BERT modified in light of NLP.

Responding to location-based triggers of cravings to return to substance use: A qualitative study

IntroductionIndividuals with substance use disorders face many challenges in establishing and maintaining recovery, most notably from contextual factors such as people, locations, events, emotions, and other triggers of cravings that could spark return to use. We sought to understand how individuals experience and cope with locations as triggers. MethodsWe conducted 31 semi-structured in-depth interviews with individuals who had recently been in substance use treatment and reported problem levels of alcohol and opioid use according to screening instruments. Interviews were designed to explore how individuals experience location-based triggers. We coded these interviews according to concepts outlined in Marlatt and Gordon's cognitive behavioral model and used thematic analysis to generate themes in participant experiences. ResultsWe found that participants described triggers as interrelated and associated with multiple people, locations, events, or emotions. Participants commonly identified ‘home’ as a trigger, one that is particularly difficult to cope with. Participants most commonly employed avoidance as a coping strategy. Finally, participants described an expectation that substance use was associated with more time spent outside of ‘home’ while recovery was associated with limited travel and more time spent at home. ConclusionsLocation-based triggers represent a particularly challenging barrier to sustained substance use disorder recovery, partly due to the amorphous and interrelated nature of such triggers. The identification of ‘home’ as a primary trigger and common use of avoidance as a coping strategy suggest the need for additional recovery support and interventions. We plan to use these findings to develop and test an intervention promoting coping strategies.

Introducing a novel dataset for facial emotion recognition and demonstrating significant enhancements in deep learning performance through pre-processing techniques

Facial expression recognition (FER) plays a pivotal role in various applications, ranging from human-computer interaction to psychoanalysis. To improve the accuracy of facial emotion recognition (FER) models, this study focuses on enhancing and augmenting FER datasets. It comprehensively analyzes the Facial Emotion Recognition dataset (FER13) to identify defects and correct misclassifications. The FER13 dataset represents a crucial resource for researchers developing Deep Learning (DL) models aimed at recognizing emotions based on facial features. Subsequently, this article develops a new facial dataset by expanding upon the original FER13 dataset. Similar to the FER + dataset, the expanded dataset incorporates a wider range of emotions while maintaining data accuracy. To further improve the dataset, it will be integrated with the extended Cohn-Kanade (CK+) dataset.This paper investigates the application of modern DL models to enhance emotion recognition in human faces. By training a new dataset, the study demonstrates significant performance gains compared with its counterparts. Furthermore, the article examines recent advances in FER technology and identifies critical requirements for DL models to overcome the inherent challenges of this task effectively. The study explores several DL architectures for emotion recognition in facial image datasets, with a particular focus on convolutional neural networks (CNNs). Our findings indicate that complex architecture, such as EfficientNetB7, outperforms other DL architectures, achieving a test accuracy of 78.9 %. Notably, the model surpassed the EfficientNet-XGBoost model, especially when used with the new dataset. Our approach leverages EfficientNetB7 as a backbone to build a model capable of efficiently recognizing emotions from facial images. Our proposed model, EfficientNetB7-CNN, achieved a peak accuracy of 81 % on the test set despite facing challenges such as GPU memory limitations. This demonstrates the model's robustness in handling complex facial expressions. Furthermore, to enhance feature extraction and attention mechanisms, we propose a new hybrid model, CBAM-4CNN, which integrates the convolutional block attention module (CBAM) with a custom 4-layer CNN architecture. The results showed that the CBAM-4CNN model outperformed existing models, achieving higher accuracy, precision, and recall metrics across multiple emotion classes. The results highlight the critical role of comprehensive and diverse data in enhancing model performance for facial emotion recognition.

Relationship between the intraindividual interplay of negative and positive exam-related emotions and the behavioral-emotional dimensions of academic procrastination

Although achievement situations trigger multiple emotions and learners experience both negative and positive emotions simultaneously, procrastination research typically examines isolated relationships between specific emotions and procrastination. The potential interplay of negative and positive emotions regarding procrastination is often overlooked, although this perspective can shed light on their dynamic relationships. Moreover, past studies primarily address the behavioral aspect of procrastination (delay) while overlooking its emotional facet (subjective discomfort), albeit this differentiation could enhance understanding of procrastination and its correlates. The present study explores the understudied interplay of the two most frequently experienced negative (anxiety) and positive (hope) exam-related emotions regarding students’ procrastination tendencies with differentiated behavioral and emotional dimensions. The study analyzes cross-sectional data from two independent studies (Nstudy1 = 266, Nstudy2 = 468) conducted in Germany among university students. Latent profile analyses (LPA) reveal three similar profiles across both studies—High Anxiety–Low Hope, Low Anxiety–High Hope and Moderate Anxiety–Moderate Hope— and two study-specific profiles namely, Low Anxiety–Moderate Hope (Study 1) and High Anxiety–Moderate Hope (Study 2). Females compared to males are more likely to belong to the anxiety-dominated profiles. Overall, students in anxiety-dominated profiles report higher scores whereas students in hope-dominated profiles report lower scores on both behavioral and emotional dimensions of academic procrastination. This study advances an integrated approach that incorporates the combined effects of both negative and positive emotions concerning procrastination. Findings further inform research efforts towards identifying risk and protective emotional factors to develop tailored intervention against academic procrastination.

A Multi-Class Facial Emotion Recognition System for Still Images 1Omage Micheal, 2Fasola Olusanjo & 3Woods Nancy C. 1,3University of Ibadan, Ibadan, Nigeria 2Federal University of Technology, Minna, Nigeria E-mails: 1micheal.omage@gmail.com, 2Sanjo.fasola@futminna.edu.ng, 3Chyn.woods@gmail.com Phone Nos: 1+2348084121913, 2+2348023138034, 3+2348037273291 ABSTRACT Facial expressions serve as a potent and natural means of human communication, conveying a wide range of emotions. In the realm of

Facial expressions serve as a potent and natural means of human communication, conveying a wide range of emotions. In the realm of artificial intelligence and computer vision, recognizing these emotions from facial cues has gained significant importance. Most research focus on detecting a single dominant emotion on any given facial image, although an image could portray several emotions. This work presents a multi-class facial emotion recognition system capable of predicting and quantifying the presence of multiple emotions in a single facial image. The system employs a Convolutional Neural Network (CNN) architecture optimized for multi-label classification across eight emotion classes. The AffectNet dataset, comprising 27,836 images with diverse ages, ethnicities, and real-world settings, was utilized for training the model, to enhance generalization capabilities. The model's performance was evaluated on an impartial benchmark dataset, FER-Plus, containing 35,710 images. The CNN model performed well, with a micro F1-score of 0.7 and a macro F1-score of 0.7 on the test set from AffectNet, and a micro F1-score of 0.7 and a macro F1-score of 0.4 on the FER-Plus dataset. Qualitative analysis demonstrated the model's capability in recognizing blended and subtle facial expressions, providing significant probability distributions across multiple emotion labels. The developed system advances the state-of-the-art in multi-label facial emotion recognition by successfully modeling emotion mixtures, contributing to a deeper understanding of this challenging task and enabling practical applications in affective computing, human-computer interaction, psychological studies, and improving accessibility and inclusivity in emotional computing technologies. Keywords: Multi-Class, Facial Emotion Recognition, Security, System, Still Images, Models, Detection Omage, M. Fasola, O. & Woods, N.C. (2024): A Multi-Class Facial Emotion Recognition System for Still Images. Journal of Advances in Mathematical & Computational Science. Vol. 12, No. 3. Pp 67-80. Available online at www.isteams.net/mathematics-computationaljournal. dx.doi.org/10.22624/AIMS/MATHS/V12N3P6

Extending speech emotion recognition systems to non-prototypical emotions using mixed-emotion model

In the conventional approach to speech emotion recognition (SER), the classifier is usually trained on acted emotional speech data to predict individual basic emotions. In this work, we extend the SER systems with the realistic assumption of the coexistence of multiple basic emotions in an utterance. We utilize the MSP-Podcast database for developing the SER system, which contains spontaneous speech utterances. From the primary and secondary emotion annotations of this database, we organize six prototypical and seven non-prototypical emotions. We then propose the mixed-emotion model and express the non-prototypical emotions as a linear combination of prototypical emotions. The combination weights of the mixed-emotion model are computed using a normalized dominance scale based algorithm inspired by the integration of basic emotion theory and dimensional emotion theory in human psychology. We first train a prototypical SER model using the ECAPA-TDNN architecture. The softmax predictions from this model serve as emotion profile inputs to the mixed-emotion model, which then predicts the non-prototypical emotions. Assuming the coexistence of multiple emotions, we only apply the utterances with uniform emotion profiles to the mixed-emotion model. The developed system continues with the conventional SER model if the emotion profile tends to delta function owing to the probable occurrence of a single prototypical emotion. We develop the proposed mixed-emotion model based SER framework using MFCC and wav2vec 2.0 extracted features. Further, we show that due to human perception variations, there exist prominent annotation variations in the non-prototypical emotion ground truths. To address that, we extend the supervised evaluation protocols in four different formulations that capture the subjective variability at different levels. The proposed system shows a best-case performance improvement of 7.10% and 8.39% over the conventional prototypical SER model for the MFCC and wav2vec 2.0 features, respectively.

Label distribution learning for compound facial expression recognition in‐the‐wild: A comparative study

AbstractHuman emotional states encompass both basic and compound facial expressions. However, current works primarily focus on basic expressions, consequently neglecting the broad spectrum of human emotions encountered in practical scenarios. Compound facial expressions involve the simultaneous manifestation of multiple emotions on an individual's face. This phenomenon reflects the complexity and richness of human states, where facial features dynamically convey a combination of feelings. This study embarks on a pioneering exploration of Compound Facial Expression Recognition (CFER), with a distinctive emphasis on leveraging the Label Distribution Learning (LDL) paradigm. This strategic application of LDL aims to address the ambiguity and complexity inherent in compound expressions, marking a significant departure from the dominant Single Label Learning (SLL) and Multi‐Label Learning (MLL) paradigms. Within this framework, we rigorously investigate the potential of LDL for a critical challenge in Facial Expression Recognition (FER): recognizing compound facial expressions in uncontrolled environments. We utilize the recently introduced RAF‐CE dataset, meticulously designed for compound expression assessment. By conducting a comprehensive comparative analysis pitting LDL against conventional SLL and MLL approaches on RAF‐CE, we aim to definitively establish LDL's superiority in handling this complex task. Furthermore, we assess the generalizability of LDL models trained on RAF‐CE by evaluating their performance on the EmotioNet and RAF‐DB Compound datasets. This demonstrates their effectiveness without domain adaptation. To solidify these findings, we conduct a comprehensive comparative analysis of 12 cutting‐edge LDL algorithms on RAF‐CE, S‐BU3DFE, and S‐JAFFE datasets, providing valuable insights into the most effective LDL techniques for FER in‐the‐wild.

Exploring Emotional Vulnerability and Sense of Agency in the Digital Realm: A Three-Year Case Study of an EFL Teacher in South Korea

This study presents the findings from a longitudinal case study on a Korean English teacher’s experience of emotional vulnerability and professional development in online teaching. Grounded in an ecological perspective on teacher agency and emotional vulnerability, the study investigates how and why the teacher exercised agency in navigating her emotional challenges and adapting to the virtual teaching environment. Data were collected from various sources, including interviews, observations, field notes, and artifacts, for three years. The findings showed that the intricate connection between the teacher’s multiple and dynamic emotions, beliefs, and concurrent working environments significantly influences her engagement in digital teaching practices. Furthermore, the teacher’s emotional vulnerability in virtual teaching facilitated her engagement in reflective and action-oriented practices, leading to a more proactive approach to exploring and implementing alternative teaching methods online. The findings highlight the interplay between emotions, agency, and identity development, emphasizing the importance of experiencing, understanding, negotiating, and expressing emotions to discover potential alternative teaching resources and methods in ever-changing teaching environments. This study provides insights into how teachers can leverage their emotions and exercise teacher agency to enhance their professional growth and adaptability in digital learning environments.

Emotional journeys: the pathways of African students to a Chinese higher education institution and performances of ‘emotion work’

ABSTRACT This article examines the emotional aspects of international student mobility (ISM) from African countries to a higher education institution in China. While there is increasing scholarly attention to the role of emotional experiences in ISM, most of this work has focused on Western destinations with a long history of education mobilities. Information on newly emerged, non-Western ISM destinations is comparatively limited. Drawing on ethnographic research of 15 months at a Chinese university, this article shows how emotions shape and re-shape pre-arrival aspirations, daily interactions, and post-graduation plans among students from African countries. Specifically, the study shows that many students confront emotional dissonance and perform diverse forms of ‘emotion work’ in attempts to find a sense of familiarity and belonging while studying abroad. This is significant as it speaks to students’ agentive subjectivity in contending with emotional challenges and navigating complex ‘emoscapes’. The paper argues that a focus on emotions contributes to a more nuanced, multifaceted understanding of contemporary educational mobilities. It illuminates how students’ multiple emotions are continuously articulated and circulated with and within these ISM flows apparently driven by economic forces in the global higher education market.

The experience of fertility concerns in patients with testicular cancer: a qualitative study

BackgroundTesticular cancer usually occurs in young adult men between the ages of 20 and 40 years, which largely coincides with the age of men’s reproductive intentions. However, a serious side effect of testicular cancer therapy could reduce the fertility of patients.PurposeTo explore the experience of fertility concerns in patients with testicular cancer.MethodsA phenomenological research was conducted on 12 patients with testicular cancer. Data collection was from May 2023 to August 2023, and Colaizzi analysis method was used to analyze the data.ResultsFour themes were found: (1) multiple worries and negative emotions, (2) fertility decision-making faces many challenges, (3) self-coping strategies for facing fertility concerns, (4) unmet supportive care needs.ConclusionMedical staff should pay attention to the fertility needs of patients with testicular cancer and provide relevant interventions and support to reduce their fertility concerns.

High-Accuracy Classification of Multiple Distinct Human Emotions Using EEG Differential Entropy Features and ResNet18

The high-accuracy detection of multiple distinct human emotions is crucial for advancing affective computing, mental health diagnostics, and human–computer interaction. The integration of deep learning networks with entropy measures holds significant potential in neuroscience and medicine, especially for analyzing EEG-based emotion states. This study proposes a method combining ResNet18 with differential entropy to identify five types of human emotions (happiness, sadness, fear, disgust, and neutral) from EEG signals. Our approach first calculates the differential entropy of EEG signals to capture the complexity and variability of the emotional states. Then, the ResNet18 network is employed to learn feature representations from the differential entropy measures, which effectively captures the intricate spatiotemporal dynamics inherent in emotional EEG patterns using residual connections. To validate the efficacy of our method, we conducted experiments on the SEED-V dataset, achieving an average accuracy of 95.61%. Our findings demonstrate that the combination of ResNet18 with differential entropy is highly effective in classifying multiple distinct human emotions from EEG signals. This method shows robust generalization and broad applicability, indicating its potential for extension to various pattern recognition tasks across different domains.

Unified in remembrance: reflections on collective war commemoration by war-affected immigrants in the Netherlands

Decades after World War II, major commemorations are still organised to collectively remember this war. Aiming for inclusive societies, the perspectives of immigrants with different war experiences are important to be heard in relation to these commemorations. Research reveals that commemorations can kindle difficult emotions such as sadness and bring distressing memories to the fore. Concurrently, psychosocial factors such as societal support and meaning making may buffer the emotional distress and contribute to a beneficial impact. It is not well known how this applies to immigrants. The present study explores how war-affected immigrants in the Netherlands relate to the Dutch Remembrance Day. Semi-structured interviews were conducted with 25 war-affected immigrants from 8 different countries (including countries in Europe, Asia, Africa, and South America). Qualitative analyses revealed that participants experienced multiple and mixed emotions during the Dutch Remembrance Day, including sadness, longing, and gratitude. Memories about own war experiences and losses often arose, of which the impact varied from distress to relief. The context of war in participants’ countries of origin, such as ongoing conflict or a contested history, often hindered commemoration of these wars. Through the Dutch commemoration, most war-affected immigrants experienced social connectedness and openness for (emotional) expression which contributed to dealing with their past experiences. Insights from this study may guide counsellors working with war-affected immigrants and help societies to organise optimally appealing commemorations within an inclusive society.

Residents’ reactance and their intervention strategies in the context of mandatory waste separation

Mandatory policies for waste separation may induce residents’ psychological reactance, which always manifests as multiple negative emotions. This psychological reaction changes dynamically and may continuously harm municipal waste management. However, how this psychological reactance changes over time and how we can intervene in it still lacks exploration. To bridge this gap, the development trends of psychological reactance for different residents and the intervention effects of information publicity were theoretically analyzed based on psychological reactance theory and persuasion theory. A four-times tracing questionnaire survey of 200 general residents from Shanghai was launched, and the data were statistically analyzed using latent variable mixed growth modeling. The results show that: (1) Mandatory policies for waste separation can lead to three trends of psychological reactance among residents: growth mode, decline mode, and inverted U mode. The latter two of these categories account for a greater proportion of the population and may be influenced by values and cultural background. (2) In all three modes, normative and knowledge-based information publicity have a mitigating effect on the growing trend of psychological reactance, while the effect on their initial state is not significant. (3) The changing trend of residents’ psychological reactance to waste separation is dynamic. Information publicity can effectively alleviate the trend of residents’ psychological reactance. The findings provide a dynamic perspective on the negative outcomes of mandatory waste separation policies and are helpful for the government to develop more appropriate policies and interventions.

Advancements in Korean Emotion Classification: A Comparative Approach Using Attention Mechanism

Recently, the analysis of emotions in social media has been considered a significant NLP task in digital and social-media-driven environments due to their pervasive influence on communication, culture, and consumer behavior. In particular, the task of Aspect-Based Emotion Analysis (ABEA), which involves analyzing the emotions of various targets within a single sentence, has drawn attention to understanding complex and sophisticated human language. However, ABEA is a challenging task in languages with limited data and complex linguistic properties, such as Korean, which follows spiral thought patterns and has agglutinative characteristics. Therefore, we propose a Korean Target-Attention-Based Emotion Classifier (KOTAC) designed to utilize target information by unveiling emotions buried within intricate Korean language patterns. In the experiment section, we compare various methods of utilizing and representing vectors of target information for the attention mechanism. Specifically, our final model, KOTAC, shows a performance enhancement on the MTME (Multiple Targets Multiple Emotions) samples, which include multiple targets and distinct emotions within a single sentence, achieving a 0.72% increase in F1 score over a baseline model without effective target utilization. This research contributes to the development of Korean language models that better reflect syntactic features by innovating methods to not only obtain but also utilize target-focused representations.

The Limits of Words: Expanding a Word-Based Emotion Analysis System with Multiple Emotion Dictionaries and the Automatic Extraction of Emotive Expressions

Wide adoption of social media has caused an explosion of information stored online, with the majority of that information containing subjective, opinionated, and emotional content produced daily by users. The field of emotion analysis has helped effectively process such human emotional expressions expressed in daily social media posts. Unfortunately, one of the greatest limitations of popular word-based emotion analysis systems has been the limited emotion vocabulary. This paper presents an attempt to extensively expand one such word-based emotion analysis system by integrating multiple emotion dictionaries and implementing an automatic extraction mechanism for emotive expressions. We first leverage diverse emotive expression dictionaries to expand the emotion lexicon of the system. To do that, we solve numerous problems with the integration of various dictionaries collected using different standards. We demonstrate the performance improvement of the system with improved accuracy and granularity of emotion classification. Furthermore, our automatic extraction mechanism facilitates the identification of novel emotive expressions in an emotion dataset, thereby enriching the depth and breadth of emotion analysis capabilities. In particular, the automatic extraction method shows promising results for applicability in further expansion of the dictionary base in the future, thus advancing the field of emotion analysis and offering new avenues for research in sentiment analysis, affective computing, and human–computer interaction.

Identification of Perceived Challenges in the Green Energy Transition by Turkish Society through Sentiment Analysis

Green energy refers to energy derived from renewable sources such as solar, wind, hydro, and biomass, which are environmentally sustainable. It aims to reduce reliance on fossil fuels and mitigate environmental impacts. In the Turkish context, alongside positive sentiments regarding the establishment of energy plants, there are also prevalent negative perspectives. Societal responses to the transition towards green energy can be effectively gauged through the analysis of individual comments. However, manually examining thousands of comments is both time-consuming and impractical. To address this challenge, this study proposes the integration of the Transformer method, a Natural Language Processing (NLP) technique. This study presents a defined NLP procedure that utilizes a multi-labeled NLP model, with a particular emphasis on the analysis of comments on social media classified as “dirty text”. The primary objective of this investigation is to ascertain the evolving perception of Turkish society regarding the transition to green energy over time and to conduct a comprehensive analysis utilizing NLP. The study utilizes a dataset that is multi-labeled, wherein emotions are not equally represented and each dataset may contain multiple emotions. Consequently, the measured accuracy rates for the risk, environment, and cost labels are, respectively, 0.950, 0.924, and 0.913, whereas the ROC AUC scores are 0.896, 0.902, and 0.923. The obtained results indicate that the developed model yielded successful outcomes. This study aims to develop a forecasting model tailored to green energy to analyze the current situation and monitor societal behavior dynamically. The central focus is on determining the reactions of Turkish society during the transition to green energy. The insights derived from the study aim to guide decision-makers in formulating policies for the transition. The research concludes with policy recommendations based on the model outputs, providing valuable insights for decision-makers in the context of the green energy transition.

Developing scale for affective response: anxiety, anger, depression

IntroductionPeople experience various negative emotions when they encounter stressful events, and these negative emotions contribute to the onset of illnesses. These emotional responses are not limited to just one; a person can experience multiple emotions at once, and the primary emotional reactions can vary depending on the severity and duration of the illness or life events. This is reason why we created a self-report scale to assess short-term emotional responses, focusing on the current emotional state experienced subjectively by patients.ObjectivesThe purpose of this study was to develop an affective response scale (ARS) and examine its validity and reliability.MethodsWe established clusters of affective via a literature review and developed preliminary items based on the structure. We conducted expert content validation to converge on the final items, followed by construct validity and reliability analyses.ResultsThe research findings indicate that the Affective Response Scale was composed of three main dimensions: anxiety, anger, and depression. Content validity results confirmed the validity of most items. The scale developed in this study was found to be valid in both exploratory and confirmatory factor analyses, and it was identified to be stable and consistent through the analysis of the internal reliability.ConclusionsThese results indicate that the ARS is highly reliable and valid, and that it can be utilized as an effective measure of the patient’s emotion and its severity.Disclosure of InterestNone Declared

Emotional signatures of climate policy support

The optimal emotional tone for climate communication has been debated by scholars and the press, but little is known about the effects of emotions on different types of policy support. In this paper we examine multiple discrete emotions people experience in reaction to climate change, and assess the strength of these emotions as predictors of support for different types of mitigation and adaptation policies. Using multi-wave, cross-sectional, nationally representative samples of American adults, we test whether guilt, anger, hope, fear, and sadness are uniquely associated with support for different types of climate policies. Guilt is most strongly related to support for personally costly policies, hope to support for proactive policies, and fear to support for regulatory policies. This research suggests that communicators should consider how their climate campaign’s emotional tone aligns with their policy goals.

The Caring Machine: Feeling AI for Customer Care

Customer care is important for its role in relationship building. This role has traditionally been performed by human customer agents; however, the emergence of interactive generative AI (GenAI) shows potential for using AI for customer care in emotionally charged interactions. Bridging practice and the academic literatures in marketing and computer science, this article develops an AI-enabled customer care journey, from accurate emotion recognition to empathetic response, emotional management support, and, finally, the establishment of an emotional connection. Marketing requirements for each of the stages are derived from in-depth interviews with top managers and a survey of chief marketing officers. By juxtaposing these requirements against the current feeling capabilities of GenAI, the authors highlight the technological challenges engineers must tackle. The article concludes with a set of marketing tenets for implementing and researching the caring machine. These include verifying emotion recognition accuracy using marketing emotion theories through multiple emotion signals and methods, utilizing prompt engineering to enhance GenAI’s emotion understanding, employing “response engineering” to personalize emotion management recommendations, and strategically deploying GenAI for emotional connection to simultaneously enhance customer emotional well-being and customer lifetime value.

MM-EMOG: Multi-Label Emotion Graph Representation for Mental Health Classification on Social Media

More than 80% of people who commit suicide disclose their intention to do so on social media. The main information we can use in social media is user-generated posts, since personal information is not always available. Identifying all possible emotions in a single textual post is crucial to detecting the user’s mental state; however, human emotions are very complex, and a single text instance likely expresses multiple emotions. This paper proposes a new multi-label emotion graph representation for social media post-based mental health classification. We first construct a word–document graph tensor to describe emotion-based contextual representation using emotion lexicons. Then, it is trained by multi-label emotions and conducts a graph propagation for harmonising heterogeneous emotional information, and is applied to a textual graph mental health classification. We perform extensive experiments on three publicly available social media mental health classification datasets, and the results show clear improvements.