Search results

With the rapid growth of the domestic lending industry, assessing whether the borrower of each loan is at risk of default is a pressing issue for financial institutions. Although there are some models that can handle such problems well, there are still some shortcomings in some aspects. The purpose of this paper is to improve the accuracy of credit assessment models.

In this paper, three different stages are used to improve the classification performance of LSTM, so that financial institutions can more accurately identify borrowers at risk of default. The first approach is to use the K-Means-SMOTE algorithm to eliminate the imbalance within the class. In the second step, ResNet is used for feature extraction, and then two-layer LSTM is used for learning to strengthen the ability of neural networks to mine and utilize deep information. Finally, the model performance is improved by using the IDWPSO algorithm for optimization when debugging the neural network.

On two unbalanced datasets (category ratios of 700:1 and 3:1 respectively), the multi-stage improved model was compared with ten other models using accuracy, precision, specificity, recall, G-measure, F-measure and the nonparametric Wilcoxon test. It was demonstrated that the multi-stage improved model showed a more significant advantage in evaluating the imbalanced credit dataset.

In this paper, the parameters of the ResNet-LSTM hybrid neural network, which can fully mine and utilize the deep information, are tuned by an innovative intelligent optimization algorithm to strengthen the classification performance of the model.

This study aims to construct a sentiment series generation method for danmu comments based on deep learning, and explore the features of sentiment series after clustering.

This study consisted of two main parts: danmu comment sentiment series generation and clustering. In the first part, the authors proposed a sentiment classification model based on BERT fine-tuning to quantify danmu comment sentiment polarity. To smooth the sentiment series, they used methods, such as comprehensive weights. In the second part, the shaped-based distance (SBD)-K-shape method was used to cluster the actual collected data.

The filtered sentiment series or curves of the microfilms on the Bilibili website could be divided into four major categories. There is an apparently stable time interval for the first three types of sentiment curves, while the fourth type of sentiment curve shows a clear trend of fluctuation in general. In addition, it was found that “disputed points” or “highlights” are likely to appear at the beginning and the climax of films, resulting in significant changes in the sentiment curves. The clustering results show a significant difference in user participation, with the second type prevailing over others.

Their sentiment classification model based on BERT fine-tuning outperformed the traditional sentiment lexicon method, which provides a reference for using deep learning as well as transfer learning for danmu comment sentiment analysis. The BERT fine-tuning–SBD-K-shape algorithm can weaken the effect of non-regular noise and temporal phase shift of danmu text.

Although many attempts to discover key segments of sport spectators have been extant, little segmentation effort has been made to reflect pandemic situations such as the COVID-19 pandemic. The purpose of this research is twofold: (1) to classify sport spectators into key segments based on perceived risks associated with a mass-gathered sporting event during the COVID-19 pandemic and (2) to identify each segment’s profiles.

Questionnaire surveys of spectators attending a Japanese rugby game during the COVID-19 pandemic (January–June 2021) were conducted (n = 1,410). A combination of hierarchical and non-hierarchical clustering methods was executed.

The results revealed the five-cluster solution as the optimal number of clusters representing the samples (i.e. spectators with extremely low-risk perception, those with low-risk perception, those with moderate-risk perception, those with high-risk perception and those with higher social risk perception). This five-cluster solution showed sufficient stability and validity. Moreover, each segment had different profiles regarding three background aspects – demographics, psychographics and behavioral variables.

This study is the first effort to segment sport spectators based on perceived risks associated with a mass-gathered sporting event in the pandemic situation. Despite extensive segmentation studies to explore sport fans, contribution reflecting the post-crisis situations is scant. Therefore, the findings provide insight into this realm by providing a new viewpoint for understanding sport spectators during a possible future pandemic era.

This study aims to investigate the customer’s willingness to participate in family takaful using the theory of interpersonal behaviour (TIB) in Pakistan.

For this study, purposive sampling was used, and 310 useable questionnaires were received from the respondents who were postgraduate students, non-users of family takaful. The respondents are residing in the largest city of Pakistan, which is Karachi. The software Analysis of Moments Structures (AMOS v.25) was used to analyse the data.

The findings of this study reveal that awareness, affect, facilitating conditions and religious obligation are pivotal in determining the customers’ intention to purchase family takaful products. In addition, perceived risk and social factors are found not to be significant predictors. Resultantly, it may also be necessary to look into the factors examined in this study and other factors that may have played a great role in the acceptance of family takaful in the case of Pakistan.

This study is limited in terms of geographic coverage as it only covers part of Karachi city as a place of investigation, and therefore, the results cannot be generalised fully. On the same note, the sampling method can also be broadened to have the actual number of respondents for generalisability purposes. Future studies may focus on the random sampling method using cluster sampling to cover other regions and provinces for a clear picture and understanding.

To the best of the authors’ knowledge, this work is one of the first studies to be carried out on empirical grounds using the TIB in the context of family takaful products in Pakistan.

The rise of communication software has changed our work style. The objectives of this study are: (1) to explore the effect of supervisors making after-hours work requests using communication software (SWRUCS) on employees’ job stress, quality of life and (2) to examine the moderating effect of personality traits and the cross-level contextual effect of social support.

A questionnaire survey was conducted to obtain information from 357 employees.

The results suggested that SWRUCS exacerbated job stress, which negatively impacted on quality of life and well-being. Moreover, different personality traits can either increase or decrease the positive or negative effect of SWRUCS on job stress. This study also revealed that social support can reduce employees’ job stress in a cross-level fashion. Furthermore, social support, especially organizational and supervisory support, can decrease the negative effect of job stress on employees’ quality of life and well-being.

Theoretically, this study has broadened the research scope of the organizational application of communication software, and practically, this study has demonstrated the reason why organizations should provide social support and select employees with suitable personality traits.

The construction industry is a primary contributor to the development of emerging economies such as the Kurdistan Region of Iraq. However, the sector is underperforming, and products are not meeting expectations. A lack of collaboration is considered a significant contributor to these issues. Various researchers have identified factors to improve collaborative approaches. However, there is still a lack of clear frameworks to help implement collaboration in the construction industry, especially in emerging economies. Therefore, this study aims to develop a framework to implement collaboration in the construction industry.

This article utilises a review of literature, questionnaire and interviews with experts in the construction industry in order to develop a framework to achieve collaboration in construction projects.

The research presents a framework that distributes the factors of collaboration over the project lifecycle stages in accordance with the Royal Institute of British Architects (RIBA) Plan of Work 2007. Each factor is divided into a set of enabling conditions which must be satisfied to ensure that the given specific factors are delivered. Additionally, the framework suggests appointing a collaboration champion at the beginning of the project to manage the process.

The research contributes to scarce literature about collaboration practices in the Kurdistan Region and in emerging economies in general.

For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the limited resources such as bandwidth and power of local devices, communication in federated learning can be much slower than in local computing. This study aims to improve communication efficiency by reducing the number of communication rounds and the size of information transmitted in each round.

This paper allows each user node to perform multiple local trainings, then upload the local model parameters to a central server. The central server updates the global model parameters by weighted averaging the parameter information. Based on this aggregation, user nodes first cluster the parameter information to be uploaded and then replace each value with the mean value of its cluster. Considering the asymmetry of the federated learning framework, adaptively select the optimal number of clusters required to compress the model information.

While maintaining the loss convergence rate similar to that of federated averaging, the test accuracy did not decrease significantly.

By compressing uplink traffic, the work can improve communication efficiency on dynamic networks with limited resources.

The purpose of the paper is to present an empirical study that examines the impact of digitalization on informative, strategic and operational marketing activities in manufacturing companies from the entrepreneurial perspective.

A research project was carried out in 205 Italian manufacturing companies by using the questionnaire method. An exploratory research study was conducted with hierarchical cluster analysis.

The analysis shows the existence of seven clusters of manufacturing companies that differ by the impact of digitalization on marketing activities from the entrepreneurial perspective. Two clusters have a high positive impact of digitalization, primarily on informative and strategic marketing activities. Two clusters are characterized by a low positive impact of digitalization and three clusters perform an intermediate level of digitalization. Furthermore, these groups of clusters differ in terms of the influence of digitalization on customer value.

The small size of the sample and the geographic origin of the companies imply limited generalizability; further research on the topic is thus recommended.

The study suggests that companies should digitalize many key marketing activities to increase marketing effectiveness and customer value. To achieve high levels of digitalization and thus increase their competitiveness, manufacturing companies should consider the importance of relevant technologies and skills.

By focussing on the impact of digitalization on informative, strategic and operational marketing, which has not yet been empirically investigated, the present study reveals many new elements concerning the marketing process in the digital era from the entrepreneur's point of view.

The purpose of this paper is to examine the purchase behavior of consumers towards green products by adapting and extending the theory of planned behavior with the inclusion of three pertinent environmental factors posited to reflect environmental consciousness in the form of environmental concern, environmental knowledge and environmental values.

The data was collected from 410 consumers at shopping malls with retail stores selling green and non-green products in a developing country using cluster sampling and analyzed using covariance-based structural equation modeling.

The findings of this study indicate that environmental factors reflecting environmental consciousness positively influence consumers’ attitude towards purchasing green products, wherein consumers’ environmental values have a stronger influence than their environmental concern and environmental knowledge. The findings also reveal that subjective norm, attitude and perceived behavioral control toward purchasing green products positively shape green purchase intention. The same positive effect is also witnessed between green purchase intention and behavior. However, perceived behavioral control towards purchasing green products had no significant influence on green purchase behavior.

This study suggests that green marketers should promote environmental consciousness among consumers to influence and shape their planned behavior towards green purchases. This could be done by prioritizing efforts and investments in inculcating environmental values, followed by enhancing environmental knowledge and finally inducing environmental concern among consumers. Green marketers can also leverage subjective norm and perceptions of behavioral control toward purchasing green products to reinforce green purchase intention, which, in turn, strengthens green purchase behavior. This green marketing strategy should also be useful to address the intention–behavior gap as seen through the null effect of perceived behavioral control on purchase behavior toward green products when this strategy is present.

This study contributes to theoretical generalizability by reaffirming the continued relevance of the theory of planned behavior in settings concerning the environment (e.g. green purchases), and theoretical extension by augmenting environmental concern, environmental knowledge and environmental values with the theory of planned behavior, resulting in an environmentally conscious theory of planned behavior. The latter is significant and noteworthy, as this study broadens the conceptualization and operationalization of environmental consciousness from a unidimensional to a multidimensional construct.

The main aim of this study is to characterise and identify specific chemo-sensory profiles of ciders from the Canary Islands (Spain).

Commercial samples of Canary ciders were compared to ciders from the Basque Country and Asturias. In total, 18 samples were studied, six for each region. The analysis comprised their sensory profiling and chemical characterisation of their polyphenolic profile, volatile composition, conventional chemical parameters and CIELAB colour coordinates. In parallel, the sensory profile of the samples from the Canary Islands was first compared with their Basque and Asturian counterparts by labelled sorting task. Then, their specific aroma profile was characterised by flash profile. Further quantification of sensory-active compounds was performed by GC–MS and GC-FID to identify the volatile compounds involved in their aroma profile.

Results show that Canary ciders present a specific chemical profile characterised by higher levels of ethanol, and hydroxycinnamic acids, mainly t-ferulic, t-coumaric and neochologenic acids, and lower levels of volatile and total acidity than their Asturian and Basque counterparts. They also present a specific aroma profile characterised by fruity aroma, mainly fruit in syrup and confectionary, and sweet flavours related to their highest levels of vinylphenols formed by transformation of hydroxycinnamic acids.

An integrated strategy to explore the typicity of the currently existing Canary ciders in the market was developed. The results are important in that they will help other regions to identify specific typical chemo-sensory profiles and to promote the creation of certifications supporting regional typicity.