Search results

With the continuous increase in Web services, efficient identification of Web services that meet developers’ needs and understanding their relationships remains a challenge. Previous research has improved recommendation effectiveness by using correlations between Web services through graph neural networks (GNNs), while it has not fully leveraged service descriptions, limiting the depth and diversity of learning. To this end, a Web services recommendation method called LLMSARec, based on Large Language Model and semantic alignment, is proposed. This study aims to extract potential semantic information from services and learn deeper relationships between services.

This method consists of two core modules: profile generation and maximizing mutual information. The profile generation module uses LLM to analyze the descriptions of services, infer and construct service profiles. Concurrently, it uses LLM as text encoders to encode inferred service profiles for enhanced service representation learning. The maximizing mutual information model aims to align the semantic features of the services text inferred by LLM with structural semantic features of the services captured by GNNs, thus achieving a more comprehensive representation of services. The aligned representation serves as an input for the model to identify services with superior matching accuracy, thereby enhancing the service recommendation capability.

Experimental comparisons and analyses were conducted on the Programmable Web platform data set, and the results demonstrated that the effectiveness of Web service recommendations can be significantly improved by using LLMSARec.

In this study, the authors propose a Web service recommendation approach based on Large Language Model and semantic alignment. By extracting latent semantic information from services and effectively aligning semantic features with structural features, new representations can be generated to significantly enhance recommendation accuracy.

A transmission tower usually experiences bolt loosening under long-term alternating cyclic load, which may lead to collapse of the tower in extreme operating conditions. The paper aims to propose a data-driven identification method for bolt looseness of complicated tower structures based on reduced-order models and numerical simulations to perceive and evaluate the health state of a tower in operation.

The equivalent stiffnesses of three types of bolt joints under various loosening scenarios are numerically determined by three-dimensional finite element (FE) simulations. The order of the FE model of a tower structure with bolt loosening is reduced by means of the component modal synthesis method, and the dynamic responses of the reducer-order model under calibration loads are simulated and used to create the dataset. An identification model for bolt looseness of the tower structure based on convolutional neural networks driven by the acceleration sensors is constructed.

An identification model for bolt looseness of the tower structure based on convolutional neural networks driven by the acceleration sensors is constructed and the applicability of the model is investigated. It is shown that the proposed method has a high identification accuracy and strong robustness to data noise and data missing. Meanwhile, the method is less dependent on the number and location of sensors and is easier to apply in real transmission lines.

This paper proposes a data-driven identification method for bolt looseness of a complicated tower structure based on reduced-order models and numerical simulations. Non-linear relationships between equivalent stiffness of bolted joints and bolt preload depicting looseness are obtained and reduced-order model of tower structure with bolt looseness is established. Finally, this paper investigates applicability of identification model for bolt looseness.

The structure and protective effect of Al-coated Nd-Fe-B magnets before and after grain boundary diffusion were studied to explore the feasibility of improving the corrosion resistance of Nd-Fe-B magnets by Al coating and Al grain boundary diffusion.

The Al coating was deposited on sintered Nd-Fe-B magnets by magnetron sputtering, and then the Al-coated Nd-Fe-B magnets were put into the vacuum tube furnace for grain boundary diffusion process. The influence of Al coating and grain boundary diffusion process on the corrosion resistance of Nd-Fe-B magnets was investigated using electrochemical tests.

Results showed that the Al coating thickness increases, the corrosion current density of Al-coated magnets first increases and then decreases with increasing coating time. The Al coating particles transform from small millet shaped particles to equiaxed polygonal particles, and finally to big millet shaped particles with increasing coating time. The diffusion temperature has little effect on the corrosion potential, and the corrosion current density of Al-diffused magnets after grain boundary diffusion is much higher than that of Al-coated magnets before grain boundary diffusion. The corrosion potential and corrosion current density of magnets first increase, and then decrease with increasing tempering temperature.

As for high intrinsic coercivity Nd-Fe-B magnets, Poor preparation of Al coatings can result in Al coatings lacking protective properties, and the Al coating should be used cautiously as the surface protective coating of magnets. Grain boundary diffusion leads to the disappearance of Al coating, and reduces the corrosion resistance of Nd-Fe-B magnets, while tempering treatment can improve the corrosion resistance of Al-diffused magnets.

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.

Learning from safety accidents and sharing safety knowledge has become an important part of accident prevention and improving construction safety management. Considering the difficulty of reusing unstructured data in the construction industry, the knowledge in it is difficult to be used directly for safety analysis. The purpose of this paper is to explore the construction of construction safety knowledge representation model and safety accident graph through deep learning methods, extract construction safety knowledge entities through BERT-BiLSTM-CRF model and propose a data management model of data–knowledge–services.

The ontology model of knowledge representation of construction safety accidents is constructed by integrating entity relation and logic evolution. Then, the database of safety incidents in the architecture, engineering and construction (AEC) industry is established based on the collected construction safety incident reports and related dispute cases. The construction method of construction safety accident knowledge graph is studied, and the precision of BERT-BiLSTM-CRF algorithm in information extraction is verified through comparative experiments. Finally, a safety accident report is used as an example to construct the AEC domain construction safety accident knowledge graph (AEC-KG), which provides visual query knowledge service and verifies the operability of knowledge management.

The experimental results show that the combined BERT-BiLSTM-CRF algorithm has a precision of 84.52%, a recall of 92.35%, and an F1 value of 88.26% in named entity recognition from the AEC domain database. The construction safety knowledge representation model and safety incident knowledge graph realize knowledge visualization.

The proposed framework provides a new knowledge management approach to improve the safety management of practitioners and also enriches the application scenarios of knowledge graph. On the one hand, it innovatively proposes a data application method and knowledge management method of safety accident report that integrates entity relationship and matter evolution logic. On the other hand, the legal adjudication dimension is innovatively added to the knowledge graph in the construction safety field as the basis for the postincident disposal measures of safety accidents, which provides reference for safety managers' decision-making in all aspects.

This paper aims to investigate the relationship linking hard technology innovation with the high-quality development (HDP) of SRDI firms. SRDI firms are typically classified as medium-sized to moderately scaled businesses renowned for their specialized, refinement, differentiation and innovation (SRDI), with a focus on providing exceptional products or services to gain a competitive advantage in specific market segments. These firms are dedicated to expanding market share and enhancing innovation capacities both locally and globally. The research also aims to scrutinize the contextual effects of digital transformation within this framework.

Hard technology innovation consists of three essential components: innovative characteristics, newly developed technology-based intellectual property rights and the volume of R&D initiatives. The evaluation of HDP was performed utilizing the entropy method, with a specific emphasis on assessing value creation and value management capabilities. Subsequently, this study explores the impact of technological innovation on the HDP of firms using a dual-dimension fixed effects model.

Every aspect of hard technology innovation is essential for promoting the HDP of businesses. The digital transformation of businesses exerts a heterogeneous moderating influence in this process. This is evident in the constructive impact on the connection between innovation attributes and the volume of fruitful R&D initiatives, as well as the HDP of firms. Conversely, the moderating effect is deemed insignificant in the association between new technology-based intellectual property and HDP.

This research delves deeper into the underlying mechanisms that underlie the promotion of HDP through hard technology innovation, thereby expanding the scope of our exploration on the HDP of SRDI firms. It establishes a theoretical framework and practical directives for achieving enhanced development quality amidst the evolving landscape of digital transformation within firms.

This study investigates the effect of auditing industry specialization (AIS) on the relative derivatives use for earnings management.

The sample chosen in this study comprises 30,599 firm-year observations of Chinese public companies from 2005 to 2018. The sample is divided into two time periods (2005–2013 and 2014–2018) according to the year when IFRS 9 was implemented (IFRS 9, first discussed by the International Accounting Standards Board in March 2008, is based on an expected credit loss model for determining new and existing expected credit losses on financial assets. The definition was completed in July 2014 and implemented in 2018). AIS was gauged with respect to audit firms and individual auditors, and measured by market share in number and scale of clients. Linear regression is adopted to test hypotheses. Moreover, two-stage least square model (2SLS) is used to eliminate the concern of possible endogeneity.

When gauged with respect to client scale, the scale-based AIS constrained the level of derivatives use for earnings management in the first period (2005–2013) while increased the level in the second period (2014–2018). The findings sustain for the analysis of audit firms and that of individual auditors, and for different definitions of AIS.

The positive AIS-IN relation after the adoption of IFRS 9 implies the sacrifice audit independence. This could be indebted to the government policy that favors local audit firms to be comparable to international Big 4 audit firms, and therefore results in competition among local auditors/audit firms in securing number rather than quality of clients.

The data of AIS in China are collected using a Python web crawler.

The purpose of this paper is to integrate job demands–resources theory with core self-evaluation theory to examine how and when sustainable project leadership stimulates sustainable project performance. Psychological empowerment is used as a mechanism between sustainable project leadership and sustainable project performance, whereas core self-evaluation acts as a moderator between sustainable project leadership and psychological empowerment.

This study adopts a time-lagged two-phase approach to collecting data from 257 software engineers working in medium-sized and large companies in the Polish financial sector.

This study reveals the direct and indirect (through psychological empowerment) effects of sustainable project leadership on sustainable project performance. It also confirms that the “sustainable project leadership–psychological empowerment” relationship strengthens with increasing level of project team members’ core self-evaluation.

This article explores a unique research framework. It significantly enriches the literature on sustainable project management, increasing the understanding of effective leadership practices that address the needs of software engineering teams to support sustainable project performance.

This study aims to identify the most significant barriers and the stationary barrier to modular construction (MC) implementation and promote MC widespread use. By doing so, the construction industry can leverage the benefits of MC, such as faster construction times, improved quality control, reduced waste and increased sustainability.

This study uses a Gini’s mean analysis approach to identify the stationary barriers hindering the MC adoption in residential projects. The research focuses on the Egyptian context and uses a questionnaire survey to gather data from professionals in the construction industry.

According to the survey findings, the top five significant MC barriers are inability to modify the design; contractors asking for high bidding prices (higher initial cost); scepticism, conservation and resistance of clients to innovation and change; transportation restrictions; and lack of a one-size-fits-all tool for the design. In addition, Gini’s mean of dispersion demonstrated that the stationary barrier that faces MC adoption is the apprehension that architectural creativity will suffer because of MC.

The identified obstacles could be useful for decision makers in countries that have not yet adopted MC and may aid in the planning process to manage the risks associated with MC projects. The paper stresses the significance of devising techniques to overcome these barriers and proposes several methods to tackle these challenges.

This study fills the knowledge gap by identifying the stationary barrier and emphasising the potential risks associated with MC barriers. Furthermore, it suggests several strategies for overcoming and reducing these barriers in developing countries residential projects.

This study presents a systematic literature review (SLR) of the interdisciplinary literature on drones in last-mile delivery (LMD) to extrapolate pertinent insights from and into the logistics management field.

Rooting their analytical categories in the LMD literature, the authors performed a deductive, theory refinement SLR on 307 interdisciplinary journal articles published during 2015–2022 to integrate this emergent phenomenon into the field.

The authors derived the potentials, challenges and solutions of drone deliveries in relation to 12 LMD criteria dispersed across four stakeholder groups: senders, receivers, regulators and societies. Relationships between these criteria were also identified.

This review contributes to logistics management by offering a current, nuanced and multifaceted discussion of drones' potential to improve the LMD process together with the challenges and solutions involved.

The authors provide logistics managers with a holistic roadmap to help them make informed decisions about adopting drones in their delivery systems. Regulators and society members also gain insights into the prospects, requirements and repercussions of drone deliveries.

This is one of the first SLRs on drone applications in LMD from a logistics management perspective.