Search results

Although numerous signal modalities are available for emotion recognition, audio and visual modalities are the most common and predominant forms for human beings to express their emotional states in daily communication. Therefore, how to achieve automatic and accurate audiovisual emotion recognition is significantly important for developing engaging and empathetic human–computer interaction environment. However, two major challenges exist in the field of audiovisual emotion recognition: (1) how to effectively capture representations of each single modality and eliminate redundant features and (2) how to efficiently integrate information from these two modalities to generate discriminative representations.

A novel key-frame extraction-based attention fusion network (KE-AFN) is proposed for audiovisual emotion recognition. KE-AFN attempts to integrate key-frame extraction with multimodal interaction and fusion to enhance audiovisual representations and reduce redundant computation, filling the research gaps of existing approaches. Specifically, the local maximum–based content analysis is designed to extract key-frames from videos for the purpose of eliminating data redundancy. Two modules, including “Multi-head Attention-based Intra-modality Interaction Module” and “Multi-head Attention-based Cross-modality Interaction Module”, are proposed to mine and capture intra- and cross-modality interactions for further reducing data redundancy and producing more powerful multimodal representations.

Extensive experiments on two benchmark datasets (i.e. RAVDESS and CMU-MOSEI) demonstrate the effectiveness and rationality of KE-AFN. Specifically, (1) KE-AFN is superior to state-of-the-art baselines for audiovisual emotion recognition. (2) Exploring the supplementary and complementary information of different modalities can provide more emotional clues for better emotion recognition. (3) The proposed key-frame extraction strategy can enhance the performance by more than 2.79 per cent on accuracy. (4) Both exploring intra- and cross-modality interactions and employing attention-based audiovisual fusion can lead to better prediction performance.

The proposed KE-AFN can support the development of engaging and empathetic human–computer interaction environment.

Groups of students were enrolled in a course that sought to produce a three-phase theoretical model over three semesters.

A design project to comprehensively address school violence was launched at a university in eastern Pennsylvania.

This article updates the recent and most critical finding of the project by illuminating specific implications of the importance of teacher training and the development toward competence in recognition of children who are emotionally and psychologically injured through proactive measures such as screening for emotional and psychological well-being.

Although the model has not been tested, screening to identify those in need of emotional support and training to support teachers is clear. Screening and training offer important opportunities to help learners build skills toward resilience to soften the effects of trauma.

A view of the “whole child” with regard to academic success could further foster social and emotional development.

Early intervention can prevent the onset of symptoms associated with posttraumatic stress and related disorders. This effort alone may significantly reduce the uncomfortable incidences and perhaps ultimate prevention of the violence that is perpetuated among children.

Preliminary research supports a continued conversation regarding effective tools to find children emotionally and psychologically at-risk, which allows teachers an opportunity for timely emotional and psychological interventions.

The purpose of this study is to explore numerical scrutinization of micropolar and Walters-B non-Newtonian fluids motion under the influence of thermal radiation and chemical reaction.

The two fluids micropolar and Walters-B liquid are considered to start flowing from the slot to the stretching sheet. A magnetic field of constant strength is imposed on their flow transversely. The problems on heat and mass transport are set up with thermal, chemical reaction, heat generation, etc. to form partial differential equations. These equations were simplified into a dimensionless form and solved using spectral homotopy analysis method (SHAM). SHAM uses the basic concept of both Chebyshev pseudospectral method and homotopy analysis method to obtain numerical computations of the problem.

The outcomes for encountered flow parameters for temperature, velocity and concentration are presented with the aid of figures. It is observed that both the velocity and angular velocity of micropolar and Walters-B and thermal boundary layers increase with increase in the thermal radiation parameter. The decrease in velocity and decrease in angular velocity occurred are a result of increase in chemical reaction. It is hoped that the present study will enhance the understanding of boundary layer flow of micropolar and Walters-B non-Newtonian fluid under the influences of thermal radiation, thermal conductivity and chemical reaction as applied in various engineering processes.

All results are presented graphically and all physical quantities are computed and tabulated.

In the era of big data, various industries are generating large amounts of text data every day. Simplifying and summarizing these data can effectively serve users and improve efficiency. Recently, zero-shot prompting in large language models (LLMs) has demonstrated remarkable performance on various language tasks. However, generating a very “concise” multi-document summary is a difficult task for it. When conciseness is specified in the zero-shot prompting, the generated multi-document summary still contains some unimportant information, even with the few-shot prompting. This paper aims to propose a LLMs prompting for multi-document summarization task.

To overcome this challenge, the authors propose chain-of-event (CoE) prompting for multi-document summarization (MDS) task. In this prompting, the authors take events as the center and propose a four-step summary reasoning process: specific event extraction; event abstraction and generalization; common event statistics; and summary generation. To further improve the performance of LLMs, the authors extend CoE prompting with the example of summary reasoning.

Summaries generated by CoE prompting are more abstractive, concise and accurate. The authors evaluate the authors’ proposed prompting on two data sets. The experimental results over ChatGLM2-6b show that the authors’ proposed CoE prompting consistently outperforms other typical promptings across all data sets.

This paper proposes CoE prompting to solve MDS tasks by the LLMs. CoE prompting can not only identify the key events but also ensure the conciseness of the summary. By this method, users can access the most relevant and important information quickly, improving their decision-making processes.

Limited research exists on the perceptions of police within specific ethnic minority groups. The primary purpose of this study is to investigate the experiences of Korean and Korean American residents in the Metro Atlanta area regarding their perceptions of cooperation with the police, particularly in relation to hate crimes, along with their perceptions of police legitimacy and other relevant factors. By focusing on this specific population, the study aims to shed light on their unique perspectives and contribute to a deeper understanding of the complex dynamics between ethnic minorities and law enforcement.

The authors’ sample comprised 128 Korean residents who were asked about their demographics, victimization experiences, self-rated English proficiency and police legitimacy. Multiple linear regression analyses were employed to investigate the impact of police legitimacy, victimization experiences and English-speaking skills on the participants' level of cooperation with the police.

Police legitimacy and self-rated levels of English proficiency emerged as the most significant factors in predicting the level of cooperation among residents with the police. Furthermore, individuals who have experienced crime victimization in the past were more willing to cooperate with the police compared to those who have not. Additionally, men showed a higher tendency to cooperate with the police compared to women participants.

The findings of this study suggest important implications for the policies and strategies aimed at enhancing the relationship between the Korean American community and the police. These implications include the need for improved language support for non-English speaking community members and the importance of building trust and fostering mutual understanding to cultivate positive police-community relations. By implementing measures based on these findings, it is recommended to promote a more inclusive and effective approach to policing within the Korean American population.

Computer vision and deep learning (DL) methods have been investigated for personal protective equipment (PPE) monitoring and detection for construction workers’ safety. However, it is still challenging to implement automated safety monitoring methods in near real time or in a time-efficient manner in real construction practices. Therefore, this study developed a novel solution to enhance the time efficiency to achieve near-real-time safety glove detection and meanwhile preserve data privacy.

The developed method comprises two primary components: (1) transfer learning methods to detect safety gloves and (2) edge computing to improve time efficiency and data privacy. To compare the developed edge computing-based method with the currently widely used cloud computing-based methods, a comprehensive comparative analysis was conducted from both the implementation and theory perspectives, providing insights into the developed approach’s performance.

Three DL models achieved mean average precision (mAP) scores ranging from 74.92% to 84.31% for safety glove detection. The other two methods by combining object detection and classification achieved mAP as 89.91% for hand detection and 100% for glove classification. From both implementation and theory perspectives, the edge computing-based method detected gloves faster than the cloud computing-based method. The edge computing-based method achieved a detection latency of 36%–68% shorter than the cloud computing-based method in the implementation perspective. The findings highlight edge computing’s potential for near-real-time detection with improved data privacy.

This study implemented and evaluated DL-based safety monitoring methods on different computing infrastructures to investigate their time efficiency. This study contributes to existing knowledge by demonstrating how edge computing can be used with DL models (without sacrificing their performance) to improve PPE-glove monitoring in a time-efficient manner as well as maintain data privacy.

Global brands have become increasingly vulnerable to external disruptions that have negative spillover effects on consumers, business and brands. This research area has recently garnered interest post-pandemic yet remains fragmented. The purpose of this paper is to recognize the most impactful exogenous brand crisis (EBC) and its affective and behavioural impact on consumers.

In Study 1, we applied repertory grid technique (RGT), photo elicitation method and ANOVA comparisons, to identify the most significant EBC, in terms of repercussions on consumer purchases. In Study 2, we performed collage construction and content analysis to ascertain the impact of the identified significant crisis (from Study 1) on consumer behaviour in terms of affective and behavioural changes.

Study 1 results reveal Spread-of-diseases and Natural disaster to be the most impactful EBC based on consumer’s purchase decisions. Study 2 findings uncover three distinct themes, namely, deviant demand, emotional upheaval and community bonding that throws light on the affective and behavioural changes in consumer behaviour during the two significant EBC events.

The collated results of the two studies draw insights towards understanding the largely unexplored conceptualisation of EBC from a multi-level (micro-meso-macro) perspective. The integrated framework drawn, highlight the roles and influences of different players in exogenous brand crisis management and suggests future research agendas based on theoretical underpinnings.

To the best of our knowledge, this is the first study which identifies the most important EBC and explicates its profound impact on consumer purchase behaviour, providing critical insights to brand managers and practitioners to take an inclusive approach towards exogenous crises.

The Astroworld concert tragedy is used as an example of crisis (mis)management and the potential utility of the 4R model. Although the 4R model has been implemented in high-risk emergency management situations, it is useful in the PR field because of its actionable approach, creating a way for practitioners to prepare for and manage crisis situations.

This is an analysis of the crisis that occurred at Astroworld, spanning preparation, day-of events, casualties and enduring reputational impact. The paper applies the 4R method to the Astroworld tragedy to show how it could have lessened or even prevented the tragedy. Finally, the SCCT model is used to explain why the official post-crisis statements were ineffective.

Social media has heightened the importance of a quick and effective organizational response to risk and crisis situations because poor responses can go viral quickly. However, social media also provides intelligence and crowd sourced information that can inform PR practitioners of emerging crisis scenarios. It is also an underutilized tool for two-way communication during crises.

The 4R approach is beneficial to general practitioners as it simplifies crisis best-practices, something essential for quick action. As our world changes and becomes less predictable, practitioners must have a clear plan to protect their organizations and the public surrounding them. This approach includes reduction, readiness, response and recovery, which are all essential in crisis communication.

The 4R method has not been explored or applied in the PR field. This paper highlights how the model has been utilized in the emergency management field and illustrates the way 4R can serve the PR field.

Automation of detecting cracked surfaces on buildings or in any industrially manufactured products is emerging nowadays. Detection of the cracked surface is a challenging task for inspectors. Image-based automatic inspection of cracks can be very effective when compared to human eye inspection. With the advancement in deep learning techniques, by utilizing these methods the authors can create automation of work in a particular sector of various industries.

In this study, an upgraded convolutional neural network-based crack detection method has been proposed. The dataset consists of 3,886 images which include cracked and non-cracked images. Further, these data have been split into training and validation data. To inspect the cracks more accurately, data augmentation was performed on the dataset, and regularization techniques have been utilized to reduce the overfitting problems. In this work, VGG19, Xception and Inception V3, along with Resnet50 V2 CNN architectures to train the data.

A comparison between the trained models has been performed and from the obtained results, Xception performs better than other algorithms with 99.54% test accuracy. The results show detecting cracked regions and firm non-cracked regions is very efficient by the Xception algorithm.

The proposed method can be way better back to an automatic inspection of cracks in buildings with different design patterns such as decorated historical monuments.

Racial stigma and racial criminalization have been centralizing pillars of the construction of Blackness in the United States. Taking such systemic injustice and racism as a given, then question then becomes how these macro-level arrangements are reflected in micro-level processes. This work uses radical interactionism and stigma theory to explore the potential implications for racialized identity construction and the development of “criminalized subjectivity” among Black undergraduate students at a predominately white university in the Midwest. I use semistructured interviews to explore the implications of racial stigma and criminalization on micro-level identity construction and how understandings of these issues can change across space and over the course of one's life. Findings demonstrate that Black university students are keenly aware of this particular stigma and its consequences in increasingly complex ways from the time they are school-aged children. They were aware of this stigma as a social fact but did not internalize it as a true reflection of themselves; said internalization was thwarted through strong self-concept and racial socialization. This increasingly complex awareness is also informed by an intersectional lens for some interviewees. I argue not only that the concept of stigma must be explicitly placed within these larger systems but also that understanding and identity-building are both rooted in ever-evolving processes of interaction and meaning-making. This research contributes to scholarship that applies a critical lens to Goffmanian stigma rooted in Black sociology and criminology and from the perspectives of the stigmatized themselves.