Search results

recent years, facial expression recognition has been widely used in human machine interaction, clinical medicine and safe driving. However, there is a limitation that conventional recurrent neural networks can only learn the time-series characteristics of expressions based on one-way propagation information.

To solve such limitation, this paper proposes a novel model based on bidirectional gated recurrent unit networks (Bi-GRUs) with two-way propagations, and the theory of identity mapping residuals is adopted to effectively prevent the problem of gradient disappearance caused by the depth of the introduced network. Since the Inception-V3 network model for spatial feature extraction has too many parameters, it is prone to overfitting during training. This paper proposes a novel facial expression recognition model to add two reduction modules to reduce parameters, so as to obtain an Inception-W network with better generalization.

Finally, the proposed model is pretrained to determine the best settings and selections. Then, the pretrained model is experimented on two facial expression data sets of CK+ and Oulu- CASIA, and the recognition performance and efficiency are compared with the existing methods. The highest recognition rate is 99.6%, which shows that the method has good recognition accuracy in a certain range.

By using the proposed model for the applications of facial expression, the high recognition accuracy and robust recognition results with lower time consumption will help to build more sophisticated applications in real world.

The research on structure function recognition mainly concentrates on identifying a specific part of academic literature and its applicability in the multidiscipline perspective. A specific part of academic literature, such as sentences, paragraphs and chapter contents are also called a level of academic literature in this paper. There are a few comparative research works on the relationship between models, disciplines and levels in the process of structure function recognition. In view of this, comparative research on structure function recognition based on deep learning has been conducted in this paper.

An experimental corpus, including the academic literature of traditional Chinese medicine, library and information science, computer science, environmental science and phytology, was constructed. Meanwhile, deep learning models such as convolutional neural networks (CNN), long and short-term memory (LSTM) and bidirectional encoder representation from transformers (BERT) were used. The comparative experiments of structure function recognition were conducted with the help of the deep learning models from the multilevel perspective.

The experimental results showed that (1) the BERT model performed best, with F1 values of 78.02, 89.41 and 94.88%, respectively at the level of sentence, paragraph and chapter content. (2) The deep learning models performed better on the academic literature of traditional Chinese medicine than on other disciplines in most cases, e.g. F1 values of CNN, LSTM and BERT, respectively arrived at 71.14, 69.96 and 78.02% at the level of sentence. (3) The deep learning models performed better at the level of chapter content than other levels, the maximum F1 values of CNN, LSTM and BERT at 91.92, 74.90 and 94.88%, respectively. Furthermore, the confusion matrix of recognition results on the academic literature was introduced to find out the reason for misrecognition.

This paper may inspire other research on structure function recognition, and provide a valuable reference for the analysis of influencing factors.

In recent years, Chinese sentiment analysis has made great progress, but the characteristics of the language itself and downstream task requirements were not explored thoroughly. It is not practical to directly migrate achievements obtained in English sentiment analysis to the analysis of Chinese because of the huge difference between the two languages.

In view of the particularity of Chinese text and the requirement of sentiment analysis, a Chinese sentiment analysis model integrating multi-granularity semantic features is proposed in this paper. This model introduces the radical and part-of-speech features based on the character and word features, with the application of bidirectional long short-term memory, attention mechanism and recurrent convolutional neural network.

The comparative experiments showed that the F1 values of this model reaches 88.28 and 84.80 per cent on the man-made dataset and the NLPECC dataset, respectively. Meanwhile, an ablation experiment was conducted to verify the effectiveness of attention mechanism, part of speech, radical, character and word factors in Chinese sentiment analysis. The performance of the proposed model exceeds that of existing models to some extent.

The academic contribution of this paper is as follows: first, in view of the particularity of Chinese texts and the requirement of sentiment analysis, this paper focuses on solving the deficiency problem of Chinese sentiment analysis under the big data context. Second, this paper borrows ideas from multiple interdisciplinary frontier theories and methods, such as information science, linguistics and artificial intelligence, which makes it innovative and comprehensive. Finally, this paper deeply integrates multi-granularity semantic features such as character, word, radical and part of speech, which further complements the theoretical framework and method system of Chinese sentiment analysis.

Previous research has little specific guidance on how to improve large infrastructures’ risk analysis. This paper aims to propose a practical risk analysis framework across the project lifecycle with Bayesian Networks (BNs).

The framework includes three phases. In the qualitative phase, primary risks were identified by literature reviews and interviews; questionnaires were used to determine key risks at each project stage and causal relationships between stage-related risks. In the quantitation, brainstorming and questionnaires, and techniques of ranked nodes/paths, risk map and Bayesian truth serum were adopted. Then, a BN-based risk assessment model was developed, and risk analysis was conducted with AgenaRisk software.

Twenty key risks across the lifecycle were determined: some risks were recurring and different risks emerged at various stages with the construction and feasibility most risky. Results showed that previous stages’ risks significantly amplified subsequent stages’ risks. Based on the causality of stage-related risks, a qualitative model was easily constructed. Ranked nodes/paths facilitated the quantification by requiring less statistical knowledge and fewer parameters than traditional BNs. As articulated by a case, this model yielded very simple and easy-to-understand representations of risks and risk propagation pathways.

Rare research has developed a BN risk assessment model from the perspective of project stages. A structured model, a propagation network among individual risks, stage-related risks, and the final adverse consequence, has been designed. This research provides practitioners with a realistic risk assessment approach and further understanding of dynamic and stage-related risks throughout large infrastructures’ lifecycle. The framework can be modified and used in other real-world risk analysis where risks are complex and develop in stages.

Managing tacit knowledge effectively and efficiently is a huge challenge for organizations. Based on the social exchange and self-determination theories, this study aims to explore the role of social interactions in motivating employees' willingness to share tacit knowledge (WSTK).

The study used a survey approach and collected data from 228 employees in service and manufacturing organizations.

Interactional justice and respectful engagement are positively related to WSTK. The perceived cost of tacit knowledge sharing (CostTKS) partially mediates the relationship between interactional justice and WSTK. Respectful engagement moderates the negative relationship between interactional justice and the perceived CostTKS.

The study advances the understanding of the role of social interaction in facilitating employee WSTK by integrating the direct and intermediate relationships involving the effect of supervisor's interactional justice and peers' respectful engagement and employee perceived CostTKS on WSTK.

The findings have important practical implications for organizations as these suggest how organizations can help tacit knowledge holders experience less negative and more supportive behaviors when they engage in voluntary TKS.

This study examines the effect of both vertical and horizontal work-related interactions on perceived CostTKS and sequentially on WSTK, thereby extending existing literature.

Food delivery apps (FDA) have been widely adopted by customers in online-to-offline (O2O) catering businesses. This study aims to explore the mechanism regarding the stickiness of FDA and indicates why customers have the intention to reuse them.

A research model was developed based on the e-commerce system successful model (ECSS model) and social influence theory. Using the data collected from 312 customers who have FDA usage experience via an online survey, the established model was empirically assessed by partial least squares based structural equation model.

The results show that factors including information quality, ease of use, convenience and various choices perceived by FDA users are significant antecedents of customer satisfaction and perceived value, which in turn positively influence customers' intention to reuse. Besides, informational social influence and normative social influence play important roles in directly or indirectly affecting customers' intention to reuse.

This study extends the e-commerce system success model and enriches the literature regarding stickiness of FDA. Besides, the understanding of social influence in FDA usage has been deepened by addressing its role in the ECSS model based on the features and contexts of such apps.

Green consumption behaviour (GCB) has been advocated to mitigate the environmental consequences of traditional consumption patterns. Besides the current circumstances, Generation Z college students are a sizable consumer group who are likely to be concerned about the future. Thus, this study aims to examine the factors affecting the college students’ GCB and the moderating effect of government support to provide new evidence from college students in China.

In addition to environmental knowledge and social media influence as the variables, government support was used as a moderator to develop the extended theory of planned behaviour (TPB) model. Purposive sampling was used to obtain 328 valid responses from Chinese college students. The collected data were analysed using partial least squares structural equation modelling.

The findings indicated that subjective norms, perceived behavioural control, environmental knowledge and social media influence substantially affect students’ GCB. Notably, the moderation analysis suggested that government support greatly strengthens the relationship between subjective norms and social media influence on the GCB of Chinese college students.

The study provides several significant practical implications as the findings could be referred by stakeholders, such as government and businesses entities, in formulating policies and strategies to encourage the consumers’ GCB in mitigating ecological consequences.

The extended TPB model that integrated environmental knowledge and social media influence with the government support as the moderator contributes to the extant literature with the evidence derived from Generation Z in China.

The purpose of this paper is to examine how crowding without violating personal space influences consumers’ channel selection and the underlying mechanism of this process. Crowded environment is ubiquitous and affects consumers’ behaviors. However, less attention has been paid to whether and how crowding influences consumers’ preference for purchasing channels.

There were three studies to test the validity of the theorized model, including two laboratory experiments and a field study. The variance analyses and mediation analyses were used to give more insights into the analytical process.

This study proposes that crowding makes consumers lose their perceived control, leading them to form certain compensatory behavior through the conversion between online and offline purchasing channels – the type of goods moderates the process of compensatory behavior.

The results of this study are helpful for retailers to design effective strategies to allocate resources into online or offline channels and to choose the appropriate types of product to promote.

Environmental clues have been widely studied in previous marketing research. Crowding, as a common environmental clue, has only been noticed in recent years. This study examines the impact of crowding on consumers’ channel preference. The results of three studies have confirmed that consumers have higher preference for offline shopping when they are in a crowded environment and found the intrinsic mechanism and the marginal scenario of this process.

The purpose of this paper is to explore the underlying mechanism of how passive social networking site (SNS) use happens from aspects of impression management concern, privacy concern, and SNS fatigue, and then examine whether sense of membership can work as a moderator in this process.

The authors proposed a research model by integrating impression management concern, privacy concern, and SNS fatigue. A total of 301 valid online questionnaires were collected, and these data were assessed by PLS-SEM.

The results show that both impression management concern and privacy concern have direct and positive effects on passive SNS use, and meanwhile they can also indirectly and positively affect passive SNS use through SNS fatigue. Besides, the relationships between impression management concern and its outcomes (SNS fatigue and passive SNS use) can be moderated by sense of membership.

This research is novel in focusing on the formation of passive SNS use and providing new insight into some factors which can trigger users’ passive behaviors in SNS usage. The findings will contribute to SNS literature by offering a well proven conceptual model that facilitates the understanding of passive SNS use.

The WeChat mini-program is a new channel for the delivery of online and mobile services, including electronic government services. Given the distinguishing characteristics and new business model of WeChat mini-programs, additional studies of mini-program-based government services are warranted. The purpose of this paper is to identify the factors that determine user adoption and usage of government WeChat mini-programs (GWMPs).

An empirical study was conducted through an online survey of Chinese GWMPs users. The proposed model was tested by analyzing the collected data using the covariance-based structural equation modeling approach.

The findings show that trust in government, trust in WeChat, trust in GWMPs and perceived convenience have significant effects on the usage of GWMPs.

This study contributes to the understanding of the GWMPs and mini-program-based government phenomenon. Theoretical implications for future e-government research as well as practical suggestions for GWMPs operators are also discussed.