Search results

This research aims to develop a robust deep-learning approach for classifying emotion in social media.

This study integrates three deep learning techniques: Bidirectional Gated Recurrent Units (BiGRU), convolutional neural networks (CNN) and an attention mechanism, resulting in the Bidirectional Gated Recurrent Units Convolution Attention (BiGRU-CNN-AT) model. The BiGRU captures potential semantic features, the CNN extracts local features and the attention mechanism identifies keywords critical for classification.

The BiGRU-CNN-AT model outperformed several state-of-the-art emotion classification algorithms. The model was compared against various baselines across multiple emotion datasets, with deep learning methods consistently surpassing traditional approaches. BiGRU and Bi-LSTM networks demonstrated superior performance, particularly when combined with attention mechanisms. Additionally, analysis of execution times indicated that the BiGRU model processed data more efficiently. They were configuring hyperparameters and integrating GloVe word embeddings, which significantly enhanced model performance, with the adam optimizer proving effective for optimization.

This paper contributes to the development of a novel framework, BiGRU-CNN-AT, which integrates bidirectional GRU, CNN and attention mechanisms for text-based emotion classification. By leveraging the strengths of each component, this framework significantly enhances accuracy in emotion classification tasks. Furthermore, the study offers comprehensive experimental analyses across multiple emotion datasets.

We aim to ascertain whether and how an individual’s social comparison affects their self-gifting motivations (SGMs).

We survey a North American sample comprising 619 Canadian and US respondents. We apply partial least squares structural equation modeling (PLS-SEM) to examine relationships between social comparison, attachment orientation, parenthood, and self-gifting motivations.

We find that social comparison positively impacts self-gifting motivations. Additionally, we find that attachment orientation and parenthood can moderate social comparison’s impact on positively valenced SGMs and negatively valenced SGMs, respectively.

We elevate and expand existing scholarship on consumers’ self-gifting. Through the current study, we contribute new, empirical evidence illuminating how individuals’ attachment orientation (i.e. secure v. insecure) and parenthood status (i.e. parent v. non-parent) serve as agency-oriented moderators to temper social comparison’s influences on SGMs.

Using data from 36,981 respondents to the Global Drug Survey (GDS) COVID-19 Special Edition, this study aims to compare changes, following the first “lockdown,” in alcohol consumption between lesbian, gay, bisexual and other sexual minority (LGB+) and heterosexual respondents with and without lifetime mental health and neurodevelopmental (MHND) conditions.

Characteristics and drinking behavior of respondents to GDS who disclosed their sexual orientation and past 30-day alcohol use were described and compared. LGB+ participants with and without MHND conditions were compared, and logistic regression models identified correlates of increased drinking among LGB+ people. The impact of changed drinking on the lives of LGB+ participants with and without MHND conditions was assessed.

LGB+ participants who reported that they were “not coping well at all” with the pandemic had twofold greater odds of reporting increased binge drinking. LGB+ participants with MHND conditions were significantly more likely than those without to report increased drinking frequency (18.7% vs 12.4%), quantity (13.8% vs 8.8%) and that changed drinking had impacted their lives.

This study, which has a uniquely large and international sample, explores aspects of alcohol use not considered in other COVID-19 alcohol use research with LGB+ people; and to the best of the authors’ knowledge, this is the first study to explore alcohol use among LGB+ people with MHND conditions.

This study aims to examine the applicability of Hartmann et al.’s (2018) service ecosystem framework to the day-to-day management of the modern sales force. The authors provide a review of the framework, acknowledging its strengths, while also indicating areas for advancement. The authors conclude with recommendations to the framework and indicate opportunities where future research could advance sales theory.

A review of the theoretical underpinnings of the service ecosystem framework is weighed against the established roles and responsibilities of the modern sales force in the literature.

The ability of the framework to capture the multi-level, multi-actor and dynamic aspects of sales represents an improvement in the conceptualization of selling is critical. Suggestions around the refinement for meso-level sales interactions and a more pliant application of service dominant-logic are offered.

The suggested extensions of the framework continue the advancement of novel theorization for the field of sales. Priorities for future research include consideration of ethical implications of the framework and formulations of new management strategies reflective of the broad and dynamic properties of the ecosystem conceptualization.

This paper provides managerial guidelines and implications tied specifically to the thick and thin crossing points and how they may impact employee decision-making.

To the best of the authors’ knowledge, this study is the first to pointedly examine the service ecosystem framework with respect to established principles of managing a modern sales force.

Practitioners and management researchers have chorused the salvific tendencies of spiritual intelligence. Whilst the emergence of spirituality and its derivatives in the workplace is widely acclaimed, the conflict that exists between spiritual ideals and the capitalist ethos of modern organisations raises questions about dark manifestations. This incongruence necessitates the consideration of the misuse of spiritual intelligence.

This paper adopts conceptual lens and theoretical arguments to interrogate the assumption of absolute constructiveness that is accorded spiritual intelligence in its framing and discusses the potential of a dark side.

The dark side of spiritual intelligence is its deployment to achieve self-serving purposes, to harm, rather than to help others. More practitioners and management researchers must acknowledge that spiritual intelligence and workplace spirituality may have dark manifestations and incorporate this reality in the assessment of organisations and the individuals within them.

This exploratory article joins the sparse extant literature on the dark side of spiritual intelligence and workplace spirituality. It contributes to the literature by offering critical insights into spiritual intelligence and the need to integrate the potential for misuse in the existing models.

In this article, we reflect on how smart technology is transforming service research discourses about service innovation and value co-creation. We adopt the concept of technology smartness’ to refer to the ability of technology to sense, adapt and learn from interactions. Accordingly, we seek to address how smart technologies (i.e. cognitive and distributed technology) can be powerful resources, capable of innovating in relation to actors’ agency, the structure of the service ecosystem and value co-creation practices.

This conceptual article integrates evidence from the existing theories with illustrative examples to advance research on service innovation and value co-creation.

Through the performative utterances of new tech words, such as onlife and materiality, this article identifies the emergence of innovative forms of agency and structure. Onlife agency entails automated, relational and performative forms, which provide for new decision-making capabilities and expanded opportunities to co-create value. Phygital materiality pertains to new structural features, comprised of new resources and contexts that have distinctive intelligence, autonomy and performativity. The dialectic between onlife agency and phygital materiality (structure) lies in the agencement of smart tech–enabled value co-creation practices based on the notion of becoming that involves not only resources but also actors and contexts.

This paper proposes a novel conceptual framework that advances a tech-based ecology for service ecosystems, in which value co-creation is enacted by the smartness of technology, which emerges through systemic and performative intra-actions between actors (onlife agency), resources and contexts (phygital materiality and structure).

The paper aims to tackle the classification of Remote Sensing Images (RSIs), which presents a significant challenge for computer algorithms due to the inherent characteristics of clustered ground objects and noisy backgrounds. Recent research typically leverages larger volume models to achieve advanced performance. However, the operating environments of remote sensing commonly cannot provide unconstrained computational and storage resources. It requires lightweight algorithms with exceptional generalization capabilities.

This study introduces an efficient knowledge distillation (KD) method to build a lightweight yet precise convolutional neural network (CNN) classifier. This method also aims to substantially decrease the training time expenses commonly linked with traditional KD techniques. This approach entails extensive alterations to both the model training framework and the distillation process, each tailored to the unique characteristics of RSIs. In particular, this study establishes a robust ensemble teacher by independently training two CNN models using a customized, efficient training algorithm. Following this, this study modifies a KD loss function to mitigate the suppression of non-target category predictions, which are essential for capturing the inter- and intra-similarity of RSIs.

This study validated the student model, termed KD-enhanced network (KDE-Net), obtained through the KD process on three benchmark RSI data sets. The KDE-Net surpasses 42 other state-of-the-art methods in the literature published from 2020 to 2023. Compared to the top-ranked method’s performance on the challenging NWPU45 data set, KDE-Net demonstrated a noticeable 0.4% increase in overall accuracy with a significant 88% reduction in parameters. Meanwhile, this study’s reformed KD framework significantly enhances the knowledge transfer speed by at least three times.

This study illustrates that the logit-based KD technique can effectively develop lightweight CNN classifiers for RSI classification without substantial sacrifices in computation and storage costs. Compared to neural architecture search or other methods aiming to provide lightweight solutions, this study’s KDE-Net, based on the inherent characteristics of RSIs, is currently more efficient in constructing accurate yet lightweight classifiers for RSI classification.

This study aims to investigate patterns of information use among undergraduate engineers as they progress through their academic programs. The primary objective was to discern how second and fourth-year students differ in their use of different types of information while performing specific tasks, namely, conducting labs, composing reports and undertaking projects.

The research used an online questionnaire to collect data, focusing on the comparative analysis of second and fourth-year engineering students’ information use. The analytical framework comprised a chi-square test, residual analysis and exploratory data analysis, for evaluating statistical significance and identifying trends over time.

The results demonstrated a statistically significant difference in information use between second and fourth year undergraduates. Notably, fourth year students exhibited a preference for disciplinary genres, such as journal articles, patents and technical reports. This coincided with a decline in fourth year students’ use of educational genres, including textbooks and instructors’ handouts, notes and slides. These shifts in information use were observed consistently across all three tasks.

The uniqueness of the study resides in its innovative approach to exploring information use by investigating the relationship between genres and tasks over the course of students’ academic programs. The research introduces a novel approach for visualizing changes in information use. By describing the evolving preferences of undergraduate students from novice to emerging professional, this study contributes valuable insights into the nuanced ways in which information is used throughout the levels of engineering education.