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.

This research examined the effect of information literacy (IL) on academic engagement (AE), cognitive engagement (CE) and academic performance among business students in Pakistan.

A cross-sectional survey was conducted to collect data from business students, recruited through a proportionate stratified convenient sampling technique, of the top 13 business institutions in Pakistan. The questionnaire was personally administered by visiting each institution with permission for data collection. A total of 554 responses were received and analyzed using the partial least squire-structural equation modeling approach.

The results exhibited that these business students perceived themselves as information literate. Furthermore, IL of business students appeared to predict positively their AE, CE and academic performance.

These results provided empirical and pragmatic insights for business educators, business librarians and accreditation bodies about IL effectiveness in academia. These findings may also inform policy and practice for IL instruction programs being carried out in business-related educational institutions not only in Pakistan but also in other countries of South Asia as they share similar characteristics.

This research would be a great contribution to the existing literature on IL, especially in the academic context as the interrelationship between IL, AE, CE and academic performance has not been investigated so far.

This study aims to examine the impact of board characteristics on firm performance while also exploring the influential mechanisms that help Chinese listed companies establish effective boards of directors and strengthen their corporate governance mechanisms.

This paper uses machine learning methods to investigate the predictive ability of the board of directors' characteristics on firm performance based on the data from Chinese A-share listed companies on the Shanghai and Shenzhen stock exchanges in China during 2008–2021. This study further analyzes board characteristics with relatively strong predictive ability and their predictive models on firm performance.

The results show that nonlinear machine learning methods are more effective than traditional linear models in analyzing the impact of board characteristics on Chinese firm performance. Among the series characteristics of the board of directors, the contribution ratio in prediction from directors compensation, director shareholding ratio, the average age of directors and directors' educational level are significant, and these characteristics have a roughly nonlinear correlation to the prediction of firm performance; the improvement of the predictive ability of board characteristics on firm performance in state-owned enterprises in China performs better than that in private enterprises.

The findings of this study provide valuable suggestions for enriching the theory of board governance, strengthening board construction and optimizing the effectiveness of board governance. Furthermore, these impacts can serve as a valuable reference for board construction and selection, aiding in the rational selection of boards to establish an efficient and high-performing board of directors.

The study findings unequivocally demonstrate the superiority of nonlinear machine learning approaches over traditional linear models in examining the relationship between board characteristics and firm performance in China. Within the suite of board characteristics, director compensation, shareholding ratio, average age and educational level are particularly noteworthy, consistently demonstrating strong, nonlinear associations with firm performance. Within the suite of board characteristics, director compensation, shareholding ratio, average age and educational level are particularly noteworthy, consistently demonstrating strong, nonlinear associations with firm performance. The study reveals that the predictive performance of board attributes is generally more robust for state-owned enterprises in China in comparison to their counterparts in the private sector.

Depression has become one of the most serious and prevalent mental health problems worldwide. The rise and popularity of social networks such as microblogs provides a wealth of psychological data for early depression detection. Language use patterns reflect emotional states and psychological traits. Differences in language use between depressed and general users may help predict and diagnose early depression. Existing work focuses on depression detection using users' social textual emotion expressions, with less psychology-related knowledge.

In this paper, we propose an RNN-capsule-based depression detection method for microblog users that improves depression detection accuracy in social texts by combining textual emotional information with knowledge related to depression pathology. Specifically, we design a multi-classification RNN capsule that enhances emotion expression features in utterances and improves classification performance of depression-related emotional features. Based on user emotion annotations over time, we use integrated learning to detect depression in a user’s social text by combining the analysis results with components such as emotion change vector, emotion causality analysis, depression lexicon and the presence of surprising emotions.

In our experiments, we test the accuracy of RNN capsules for emotion classification tasks and then validate the effectiveness of different depression detection components. Finally, we achieved 83% depression detection accuracy on real datasets.

The paper overcomes the limitations of social text-based depression detection by incorporating more psychological background knowledge to enhance the early detection success rate of depression.

Social recommendation has been recognized as a kind of e-commerce with large potential, but how social recommendations influence consumer decisions is still unclear. This paper aims to investigate how recommendations from different social ties influence consumers’ purchase intentions through both behavior and brain activity.

Utilizing behavioral (N = 70) and electroencephalogram (EEG) (N = 49) experiments, this study explored participants’ behavior and brain responses after being recommended by different social ties. The data were analyzed using statistical inference and event-related potential (ERP) analysis.

Behavioral results show that social tie strength positively impacts purchase intention, which can be fitted by a logarithmic model. Moreover, recommender-to-customer similarity and product affect mediate the effect of tie strength on purchase intention serially. EEG findings show that recommendations from weak tie strength elicit larger N100, N200 and P300 amplitudes than those from strong tie strength. These results imply that weak tie strength may motivate individuals to recruit more mental resources in social recommendation, including unconscious processing of consumer attention and conscious processing of cognitive conflict and negative emotion.

This study considers the effects of continuous social ties on purchase intention and models them mathematically, exploring the intrinsic mechanisms by which strong and weak ties influence purchase intentions through recommender-to-customer similarity and product affect, contributing to the applications of the stimulus-organism-response (SOR) model in the field of social recommendation. Furthermore, our study adopting EEG techniques bridges the gap of relying solely on self-report by providing an avenue to obtain relatively objective findings about the consumers’ early-occurred (unconscious) attentional responses and late-occurred (conscious) cognitive and emotional responses in purchase decisions.

Although online reviews have become a key source of information for consumer purchasing decisions, little is known about how the concreteness of language used in these reviews influences perceptions of deception. This study aims to address this important gap by drawing on psycholinguistic research and Language Expectancy Theory to examine how and when the concreteness of online reviews (abstract vs concrete) impacts consumers’ perceived deception.

Two scenario-based experiments were conducted to examine how the concreteness of online reviews (abstract vs concrete) influences consumers’ perceptions of deception, considering the mediating role of psychological distance to online reviews and the moderating effects of Machiavellianism (Mach) and reviewer identity disclosure.

Online reviews that include concrete language lead to lower perceived deception by reducing consumers’ psychological distance from the review. For consumers with higher levels of Mach, online reviews written in abstract (vs concrete) language result in higher perceived deception via psychological distance, while for consumers with lower Mach, online reviews written in concrete (vs abstract) language result in higher perceived deception via psychological distance.

To the best of the authors’ knowledge, this study is one of the first to highlight the relevance of linguistic style (i.e. concrete review vs abstract review) on consumers’ perceived deception toward online reviews in the context of e-commerce.

The framework enables managers of online retailing platforms to identify the most effective strategies to decrease consumers’ perceived deception via the appropriate utilize of linguistic styles of online reviews.

This study contributes to both theory and practice by deepening knowledge of how and when the concreteness of online reviews (abstract vs concrete) affects consumers’ perceived deception and by helping managers of online retailing platforms make the most effective\ strategies for reducing consumers’ perceived deception toward online reviews during online shopping.

Microblogs are communication platforms for companies and consumers that challenge companies' brand marketing strategies. This paper provides a theoretical basis for expanding microblog applications and a practical basis for improving the effectiveness of brand marketing.

The authors use factor analysis to extract the factors of microblog user influence and construct a structural equation model to reveal the interaction mechanism of the influencing factors. Additionally, the authors clarify the promotion and enhancement effects of these factors.

Microblog user influence can be converted into richness, interaction and value factors. The richness factor significantly affects the latter two, whereas the interaction factor does not affect the value factor.

First, the sample used is limited to media industry practitioners. To increase generalizability, diverse groups should be included in future studies. Second, this model's theoretical explanatory ability can be further developed by adding other meaningful factors beyond the existing ones.

This study analyzes the factors of microblog user influence in China and validates the relevant elements. As a result, it improves the influence research on social media users and benefits the practice of information recommendation and microblog marketing.

This paper aims to analyze brand loyalty and electronic word-of-mouth (eWOM) antecedents in restaurants: (1) utilitarian and hedonic benefits, (2) brand satisfaction and (3) brand love. It also provides valuable knowledge through the comparison between positive and negative restaurant experiences.

A survey was carried out of restaurant satisfied and dissatisfied consumers. Structural equation modeling (SEM) and multi-group analysis (MGA) were performed to examine the cause-and-effect relationship in both groups.

The results show the relevance of benefits, brand satisfaction and brand love as causes for brand loyalty and e-WOM. Also, these relationships are significantly stronger for dissatisfied consumers than for satisfied ones.

The outcome of the research provides new insights to develop a conceptual stimulus-organism-response (S-O-R) model of consumers’ restaurant behavior by drawing comparisons across satisfied and dissatisfied ones.

Current catch documentation and traceability practices in the Philippines are paper-based. However, with technological advancements, a shift toward electronic documentation has become a global trend to combat illegal, unreported and unregulated (IUU) fishing. This study aims to determine the factors influencing fishers' preference for a mobile traceability platform and identify the challenges in achieving a digital tuna supply chain in Davao Region, Philippines.

A survey of 178 tuna fishers was conducted in select sites in Davao Region using a semi-structured questionnaire. Factors influencing fishers' preference for a mobile traceability platform were identified using logistic regression.

Results revealed that one-third of the fishers (34.5%) preferred a mobile traceability platform. Membership in organizations and higher educational attainment increased the preference for a mobile traceability platform. Meanwhile, respondents' knowledge was associated with a preference for paper-based traceability. This association between knowledge and preference was in the context of catch recording performed by government agencies and fisherfolk associations in landing sites. Intensified support aimed at increasing the fishers' literacy and access to technological devices, including the internet and smartphones, is emphasized to provide them with the basic requirements for participating in mobile traceability systems. Moreover, several challenges in implementing digital traceability beyond fisherfolk were identified.

This study amplifies the need for infrastructure and legislation to support the implementation of a digital tuna supply chain and eliminate IUU fishing.

This study aims to develop an electrochemical sensor for the detection of benzophenone (BP) as an alternative to conventional techniques that are known, expensive, complex and less sensitive.

The developed sensor is a platinum electrode modified with a plasticized polymer film based on ß-cyclodextrin, using PVC as the polymer, PEG as the plasticizer and ß-CD as the ionophore. This sensor is characterized by various techniques, such as optical microscopy, scanning electron microscopy and cyclic voltammetry. This latter is also used for analyzing kinetic processes at the electrode/electrolyte interface and to evaluate the selectivity and sensitivity of the sensor.

The results highlight the performance of our sensor. In fact, it exhibits a linear response extending from 10⁻¹⁹ to 10⁻¹³ M, with a correlation coefficient of 0.9836. What is more, it has an excellent detection limit of 10⁻¹⁹ M and a good sensitivity of 21.24 µA/M.

The results of this investigation demonstrated that the developed sensor is an analytical tool of choice for the monitoring of BP in the aqueous phase. The suggested sensor is fast, simple, reproducible and inexpensive.