Search results

To predict the price of battery-grade lithium carbonate accurately and provide proper guidance to investors, a method called MFTBGAM is proposed in this study. This method integrates textual and numerical information using TCN-BiGRU–Attention.

The Word2Vec model is initially employed to process the gathered textual data concerning battery-grade lithium carbonate. Subsequently, a dual-channel text-numerical extraction model, integrating TCN and BiGRU, is constructed to extract textual and numerical features separately. Following this, the attention mechanism is applied to extract fusion features from the textual and numerical data. Finally, the market price prediction results for battery-grade lithium carbonate are calculated and outputted using the fully connected layer.

Experiments in this study are carried out using datasets consisting of news and investor commentary. The findings reveal that the MFTBGAM model exhibits superior performance compared to alternative models, showing its efficacy in precisely forecasting the future market price of battery-grade lithium carbonate.

The dataset analyzed in this study spans from 2020 to 2023, and thus, the forecast results are specifically relevant to this timeframe. Altering the sample data would necessitate repetition of the experimental process, resulting in different outcomes. Furthermore, recognizing that raw data might include noise and irrelevant information, future endeavors will explore efficient data preprocessing techniques to mitigate such issues, thereby enhancing the model’s predictive capabilities in long-term forecasting tasks.

The price prediction model serves as a valuable tool for investors in the battery-grade lithium carbonate industry, facilitating informed investment decisions. By using the results of price prediction, investors can discern opportune moments for investment. Moreover, this study utilizes two distinct types of text information – news and investor comments – as independent sources of textual data input. This approach provides investors with a more precise and comprehensive understanding of market dynamics.

We propose a novel price prediction method based on TCN-BiGRU Attention for “text-numerical” information fusion. We separately use two types of textual information, news and investor comments, for prediction to enhance the model's effectiveness and generalization ability. Additionally, we utilize news datasets including both titles and content to improve the accuracy of battery-grade lithium carbonate market price predictions.

When recommending products to consumers, it is important to be able to accurately predict how consumers will rate them. However, existing collaborative filtering models are difficult to achieve a balance between rating prediction accuracy and complexity. Therefore, the purpose of this paper is to propose an accurate and effective model to predict users’ ratings of products for the accurate recommendation of products to users.

First, we introduce an attention mechanism that dynamically assigns weights to user preferences, highlighting key interaction information and enhancing the model’s understanding of user behavior. Second, a fold embedding strategy is employed to segment user interaction data, increasing the information density of each subset while reducing the complexity of the attention mechanism. Finally, a masking strategy is integrated to mitigate overfitting by concealing portions of user-item interactions, thereby improving the model’s generalization ability.

The experimental results demonstrate that the proposed model significantly minimizes prediction error across five real-world datasets. On average, the evaluation metrics root mean square error (RMSE) and mean absolute error (MAE) are reduced by 9.11 and 13.3%, respectively. Additionally, the Friedman test results confirm that these improvements are statistically significant. Consequently, the proposed model more accurately captures the intrinsic correlation between users and products, leading to a substantial reduction in prediction error.

We propose a novel collaborative filtering model to learn the user-item interaction matrix effectively. Additionally, we introduce a fold embedding strategy to reduce the computational resource consumption of the attention mechanism. Finally, we implement a masking strategy to encourage the model to focus on key features and patterns, thereby mitigating overfitting.

The existing research on institutional distance implicitly posits the monotonic effect of contextual differences on the multinational enterprise (MNE) behaviors (e.g. entry mode, research and development (R&D) investment and subsidiary reverse knowledge transfer). Namely, MNEs from the same home to the same host countries are thought to have homogenous perceptions on the institutional influences and thus behave similarly. However, the authors argue that MNEs, due to their different performance aspirations in host countries, will have heterogenous perceptions on such contextual influences and thereafter behave differently.

Drawing on the behavioral theory of the firm and employing a unique sample comprised of 140 Chinese MNEs' foreign direct investments (FDIs) in Taiwan in 2017, the authors developed and tested the hypotheses.

The authors found that the emerging-market MNEs' (EMNEs’) perceptions of higher local institutional difficulties will be strengthened when their local performances are below their aspiration levels, making them more risk-taking. Nevertheless, EMNEs' local experiences and local equity-based partnerships will mitigate such negative perceptions, mitigating their risk-taking orientation.

The empirical findings make contributes to the international business (IB) literature by extending knowledge on the determinants and conditions of the heterogeneity in EMNEs' behavioral orientations when in face of the same institutional distance. The authors also provide managerial implications by showing that EMNEs' firm-specific resources (i.e. local experience and local equity-based partnership) will alter their perceptions of local institutional difficulties, leading to different behavioral orientations.

Information is presented in various modalities such as text and images, and it can quickly and widely spread on social networks and among the general public through key communication nodes involved in public opinion events. Therefore, by tracking and identifying key nodes of public opinion, we can determine the direction of public opinion evolution and timely and effectively control public opinion events or curb the spread of false information.

This paper introduces a novel multimodal semantic enhanced representation based on multianchor mapping semantic community (MAMSC) for identifying key nodes in public opinion. MAMSC consists of four core components: multimodal data feature extraction module, feature vector dimensionality reduction module, semantic enhanced representation module and semantic community (SC) recognition module. On this basis, we combine the method of community discovery in complex networks to analyze the aggregation characteristics of different semantic anchors and construct a three-layer network module for public opinion node recognition in the SC with strong, medium and weak associations.

The experimental results show that compared with its variants and the baseline models, the MAMSC model has better recognition accuracy. This study also provides more systematic, forward-looking and scientific decision-making support for controlling public opinion and curbing the spread of false information.

We creatively combine the construction of variant autoencoder with multianchor mapping to enhance semantic representation and construct a three-layer network module for public opinion node recognition in the SC with strong, medium and weak associations. On this basis, our constructed MAMSC model achieved the best results compared to the baseline models and ablation evaluation models, with a precision of 91.21%.

This paper aims to investigate the relationship between corporate environmental, social and governance (ESG) ratings and leverage manipulation and the moderating effects of internal and external supervision.

The authors draw on a sample of Chinese non-financial A-share-listed firms from 2013 to 2020 to explore the effect of ESG ratings on leverage manipulation. Robustness and endogeneity tests confirm the validity of the regression results.

ESG ratings inhibit leverage manipulation by improving social reputation, information transparency and financing constraints. This effect is weakened by internal supervision, captured by the ratio of institutional investor ownership, and strengthened by external supervision, captured by the level of marketization. The effect is stronger in non-state-owned firms and firms in non-polluting industries. The governance dimension of ESG exhibits the strongest effect, with comprehensive environmental governance ratings and social governance ratings also suppressing leverage manipulation.

Firms should strive to cultivate environmental awareness, fulfil their social responsibilities and enhance internal governance, which may help to strengthen the firm’s sustainability orientation, mitigate opportunistic behaviours and ultimately contribute to high-quality firm development. The top managers of firms should exercise self-restraint and take the initiative to reduce leverage manipulation by establishing an appropriate governance structure and sustainable business operation system that incorporate environmental and social governance in addition to general governance.

Policymakers and regulators should formulate unified guidelines with comprehensive criteria to improve the scope and quality of ESG information disclosure and provide specific guidance on ESG practice for firms. Investors should incorporate ESG ratings into their investment decision framework to lower their portfolio risk.

This study contributes to the literature in four ways. Firstly, to the best of the authors’ knowledge, it is among the first to show that high ESG ratings may mitigate firms’ opportunistic behaviours. Secondly, it identifies the governance factor of leverage manipulation from the perspective of firms’ subjective sustainability orientation. Thirdly, it demonstrates that the relationship between ESG ratings and leverage manipulation varies with the level of internal and external supervision. Finally, it highlights the importance of governance in guaranteeing the other two dimensions’ roles by decomposing overall ESG.

With increasing complexity of construction projects and new construction processes and methods are adopted, more safety hazards are emerging at construction sites, requiring the application of the modern risk management methods. As an emerging technology, digital twin has already made valuable contributions to safety risk management in many fields. Therefore, exploring the application of digital twin technology in construction safety risk management is of great significance. The purpose of this study is to explore the current research status and application potential of digital twin technology in construction safety risk management.

This study followed a four-stage literature processing approach as outlined in the systematic literature review procedure guidelines. It then combined the quantitative analysis tools and qualitative analysis methods to organize and summarize the current research status of digital twin technology in the field of construction safety risk management, analyze the application of digital twin technology in construction safety risk management and identify future research trends.

The research findings indicate that the application of digital twin technology in the field of construction safety risk management is still in its early stages. Based on the results of the literature analysis, this paper summarizes five aspects of digital twin technology's application in construction safety risk management: real-time monitoring and early warning, safety risk prediction and assessment, accident simulation and emergency response, safety risk management decision support and safety training and education. It also proposes future research trends based on the current research challenges.

This study provides valuable references for the extended application of digital twin technology and offers a new perspective and approach for modern construction safety risk management. It contributes to the enhancement of the theoretical framework for construction safety risk management and the improvement of on-site construction safety.

The purpose of this study is to investigate the potential of using computer vision and deep learning (DL) techniques for improving safety on construction sites. It provides an overview of the current state of research in the field of construction site safety (CSS) management using these technologies. Specifically, the study focuses on identifying hazards and monitoring the usage of personal protective equipment (PPE) on construction sites. The findings highlight the potential of computer vision and DL to enhance safety management in the construction industry.

The study involves a scientometric analysis of the current direction for using computer vision and DL for CSS management. The analysis reviews relevant studies, their methods, results and limitations, providing insights into the state of research in this area.

The study finds that computer vision and DL techniques can be effective for enhancing safety management in the construction industry. The potential of these technologies is specifically highlighted for identifying hazards and monitoring PPE usage on construction sites. The findings suggest that the use of these technologies can significantly reduce accidents and injuries on construction sites.

This study provides valuable insights into the potential of computer vision and DL techniques for improving safety management in the construction industry. The findings can help construction companies adopt innovative technologies to reduce the number of accidents and injuries on construction sites. The study also identifies areas for future research in this field, highlighting the need for further investigation into the use of these technologies for CSS management.

Body language is an integral part of interpersonal communication and exchange, which can convey rich emotions, intentions and information. However, how anchor’s body language works in live-streaming e-commerce (LSE) has yet to receive adequate attention. Based on dual systems theory of decision-making, this paper aims to explore the impact of anchor’s body language on the performance of LSE from the perspective of customer engagement behavior and to examine the moderating role of anchor’s relational social interaction.

The authors confirmed the theoretical model through empirical analysis of structured data from 1,415 actual livestreaming rooms from Douyin, as well as unstructured data of 418,939 min of video and audio, 1,985,473 words of text and 423,302 keyframe images.

The study found that anchor’s body language has a significant positive effect on the performance of LSE, and customer engagement behavior plays a partially mediating role. The moderating effect suggests that anchor’s relational social interaction and body language have substitution effects in enhancing customer engagement behavior and the performance of LSE, which reveals the substitution relationship between anchor’s verbal and nonverbal interactions in LSE.

This study is one of the earlier literature focusing on anchor’s body language, and the findings provide practical references for enhancing customer engagement behavior and achieving performance growth in LSE.

This study uses the approach-avoidance motivation theory (AAMT) to investigate how various influence tactics used by broadcasters affect consumers’ promotion focus and purchase intention in the context of live-streaming social commerce. This study also examines the moderating role of broadcasters’ physical appearance in the relationship between consumer promotion focus and purchase intention.

Data were collected from 810 consumers on the Taobao live-streaming platform in China. Multivariate techniques and structural equation modeling were used to analyze data and test the research hypotheses.

Results indicate that both rational and emotional influence tactics positively influence consumers’ promotion focus and purchase intention. However, the influences of coercive tactics are mixed: while promise tactics positively influence consumers’ promotion focus and purchase intention, threat tactics have a negative affect. Furthermore, broadcasters’ physical appearance positively moderates the relationship between consumers’ promotion focus and purchase intention.

This study extends the AAMT by proposing and empirically testing a novel model that explains the effects of broadcasters’ influence tactics and physical appearance on consumer behavior within live-streaming social commerce. The findings provide new insights into how broadcasters’ strategies and attributes drive consumers’ positive motivation and behaviors. These insights are valuable for academic researchers and business managers in making informed decisions about recruiting and using broadcasters to achieve favorable consumer outcomes in live-streaming social commerce.

To incentivize innovation, support competitiveness, lower skill scarcities, and alleviate the housing affordability difficulty, proponents underscore the pertinence of embracing contemporary construction methodologies, with particular emphasis on volumetric modular construction (VMC) as a sustainable paradigm for production and consumption. However, construction industry stakeholders in Australia have encountered profound challenges in adopting VMC, as its adoption remains significantly low. Therefore, this study investigated the constraints that hinder VMC in the Australian construction industry.

The study used qualitative methodology using semi-structured interviews as a core approach to glean professional experts' perspectives and insights, along with Pareto and mean index score analyses.

The study identified 77 reported and validated VMC constraints by professionals, categorizing them into eight categories: cultural, economic, knowledge, market, regulatory, stakeholder, supply chain, and technological. The mean index score analysis reveals stakeholder (µ = 9.67) constraints are the most significant, followed by cultural (µ = 9.62) and regulatory (µ = 9.11) constraints. Pareto analysis revealed 25 of the 77 constraints as ‘vital few” among different categories. This study presented causal relationships and mitigation strategies for VMC constraints, followed by an argument on whether VMC adoption in Australia requires a nudge or mandate.

This study offers guidance for efficient resource allocation, aiding management and government policy formulation. It's also valuable for global audiences, especially countries transitioning to modular construction.

This is one of the first studies to identify VMC constraints and delineate them into different categories in Australia, identify their causal interrelationships, and deliver countermeasures to overcome them.