Search results

Previous research mainly uses graph neural networks on syntactic dependency graphs, often neglecting emotional cues in sarcasm detection and failing to integrate image features for multimodal information effectively. To address these limitations, this study proposes a novel multimodal sarcasm detection model based on the directed graph isomorphism network with sentiment enhancement and multimodal fusion (DGIN-SE-MF).

The approach extracts image and text features through vision transformer and BERT, respectively. To deeply integrate the extracted features, the author develops a text-guided multi-head attention fusion mechanism module. Subsequently, a directed graph is constructed through SE and the multimodal factorized bilinear pooling method to integrate image features into the graph. The DGIN then fuses the image and text features, using a weighted attention mechanism to generate the final representation.

The model is validated on three datasets: English, Chinese and an Indonesian–English dataset. The results demonstrate that the proposed model consistently outperforms other baseline models, particularly on the Chinese and English sarcasm datasets, achieving F1 scores of 88.75 % and 83.10 %, respectively.

The proposed model addresses the inadequacies of previous methods by effectively integrating emotional cues and image features into sarcasm detection. To the best of the authors’ knowledge, this is the first work to leverage a DGIN-SE-MF for this task, leading to significant improvements in detection performance across different languages.

Lean Six Sigma (LSS) implementation follows a structured approach called define-measure-analyze-improve-control (DMAIC). Earlier research about its application in emergency healthcare services shows that it requires organizational transformation, which many healthcare setups find difficult. The Kotter change management model facilitates organizational transformation but has not been attempted in LSS settings till now. This study aims to integrate the LSS framework with the Kotter change management model to come up with an integrated framework that will facilitate LSS deployment in emergency health services.

Two-stage Delphi method was conducted by using a literature review. First, the success factors and barriers of LSS are investigated, especially from an emergency healthcare point of view. The features and benefits of Kotter's change management models are then reviewed. Subsequently, they are integrated to form a framework specific to LSS deployment in an emergency healthcare set-up. The elements of this framework are analyzed using expert opinion ratings. A new framework for LSS deployment in emergency healthcare has been developed, which can prevent failures due to challenges faced by organizations in overcoming resistance to changes.

The eight steps of the Kotter model such as establishing a sense of urgency, forming a powerful guiding coalition, creating a vision, communicating the vision, empowering others to act on the vision, planning for and creating short-term wins, consolidating improvements and producing still more change, institutionalizing new approaches are derived from the eight common errors that managers make while implementing change in the institution. The study integrated LSS principles and Kotter’s change management model to apply in emergency care units in order to reduce waste and raise the level of service quality provided by healthcare companies.

The present study could contribute knowledge to the literature by providing a framework to integrate lean management and Kotter's change management model for the emergency care unit of the healthcare organization. This framework guides decision-makers and organizations as proper strategies are required for applying lean management practices in any system.

The proposed framework is unique and no other study has prescribed any integrated framework for LSS implementation in emergency healthcare that overcomes resistance to change.

The purpose of this study was to evaluate the environmental footprint of a university of applied sciences in 2019 and 2020, including the effects of the lockdown periods. The study identified the main sources of emissions and assessed the pandemic-related effects.

Using the life cycle assessment methodology, this study analysed the university’s direct and indirect emissions during a regular year of operation (2019) and compared them with those generated during the lockdown periods in 2020. For the activity areas mobility, energy, waste, IT and paper, gastronomy and water, specific, primary bottom-up inventory data were gathered before and during the pandemic. The data were assessed with 15 environmental impact assessment methods of the environmental footprint framework.

The results of a regular year of operation (2019) depicted that student and employee commuting and business travel contributed with 86% largely to the total global warming potential of 2,572 t CO₂-eq. The pandemic-induced changes in commuting and business travel resulted in a 60% reduction, leading to a drop to 1,075 t CO₂-eq (2020). In contrast, the environmental footprint due to energy consumption remained almost on the same level, irrespective of the absences on-site in 2020.

This study has the potential to shape post-pandemic environmental efforts and policies in higher education institutions and contribute to a much-needed baseline against which mitigation efforts can be compared with. Unlike other studies, this study goes beyond the carbon footprint, expanding the discussion to additional environmental and human health impact categories by applying the environmental footprint framework.