Fusing hybrid attentive network with self-supervised dual-channel heterogeneous graph for knowledge tracing

Recently the large-scale influence of Covid-19 promoted the fast development of intelligent tutoring systems (ITS). As a major task of ITS, Knowledge Tracing (KT) aims to capture a student's dynamic knowledge state based on his historical response sequences and provide personalized learning assistance to him. However, most existing KT methods have encountered the data sparsity problem. In real scenarios, an online tutoring system usually has an extensive collection of questions while each student can only interact with a limited number of questions. As a result, the records of some questions could be extremely sparse, which degrades the performance of traditional KT models. To resolve this issue, we propose a Dual-channel Heterogeneous Graph Network (DHGN) to learn informative representations of questions from students' records by capturing both the high-order heterogeneous and local relations. As the supervised learning manner applied in previous methods is incapable of exploiting unobserved relations between questions, we innovatively integrate a self-supervised framework into the KT task and employ contrastive learning via the two channels of DHGN, supplementing as an auxiliary task to improve the KT performance. Moreover, we adopt the attention mechanism, which has achieved impressive performance in natural language processing tasks, to effectively capture students' knowledge state. But the standard attention network is inapplicable to the KT task because the current knowledge state of a student usually shows strong dependency on his recently interactive questions, unlike the situation of language processing tasks, which focus more on the long-term dependency. To avoid the inefficiency of standard attention networks in the KT task, we further devise a novel Hybrid Attentive Network (HAN), which produces both the global attention and the hierarchical local attention to model the long-term and short-term intents, respectively. Then, by the gating network, a student's long-term and short-term intents are combined for efficient prediction. We conduct extensive experiments on several real-world datasets. Experimental results demonstrate that our proposed methods achieve significant performance improvement compared to existing state-of-the-art baselines, which validates the effectiveness of the proposed dual-channel heterogeneous graph framework and hybrid attentive network. © 2023 Elsevier Ltd

Advertising strategy and channel structure selection on an online retail platform

PurposeThis paper considers a supply chain with a manufacturer (she) selling through an online retail platform (he) and studies the channel structure choices of two firms when investing in advertising.Design/methodology/approachThe authors assume that the platform provides the manufacturer with an agency and/or reselling channel;thus, there are three possible channel structures: agency channel, reselling channel and dual channel. By developing a game-theoretic model, the authors investigate the channel structure choices of two firms when advertising separately, simultaneously and cooperatively and analyze the optimal combination strategy of channel structure and advertising scheme for both firms.FindingsWhen the advertising efforts of the two firms are independent of each other, the equilibrium results show that different advertising schemes lead to different channel choices. For the manufacturer, it is optimal to choose the dual channel structure and adopt the advertising scheme that both subsidizes platform advertising and advertises on her own. For the platform, this combination is also optimal at a high commission rate;otherwise, the advertising scheme in which both firms advertise simultaneously is optimal and he is better off switching from the dual channel structure to the reselling channel structure as interchannel substitution intensity increases. The above results still hold for complementary advertising efforts and asymmetric marginal advertising costs, while in the case of substitutable advertising efforts, one firm may ride on another firm's advertising efforts, leading to different strategic combinations.Originality/valueThis paper not only provides useful guidance for manufacturers and platforms in channel selection and advertising strategy, but also theoretically enriches the literature on manufacturer encroachment.

Consumers' purchasing decisions of a dual-channel supply chain system under return and warranty policies

With the rapid growth of internet technology, particularly during the COVID-19 pandemic, a major portion of consumers have intended to do online shopping. Also the use of eco-friendly products is essential in today's environment and human health. Thus this paper investigates the consumers' purchasing behaviour towards substitutable non-green and eco-friendly products in a dual-channel (offline and online channels) supply chain system. The manufacturer offers a novel combination of promotions such as a return policy in the online channel and a warranty policy in the offline channel depending on the position and situation of consumers. Here, the environmental burden is reduced by considering remanufacturing/refurbishing used products during the warranty period. Therefore, consumers' demand depends on price, eco-friendliness level, warranty and return agreements. The entire problem is modelled under centralised and decentralised decision-making scenarios. Finally, the profit maximisation problem is formulated and solved in the game theory framework. A series of sensitivity analyses of various parameters is conducted numerically to validate the problem. It is observed that, due to the instant return facility in the online channel, online demand is higher than the offline with a higher warranty period. © 2023 Informa UK Limited, trading as Taylor & Francis Group.

Effects of COVID-19 outbreak in pricing and collaboration of a health-social dual-channel supply chain.

The coronavirus pandemic has affected most aspects of product supply and consumer behaviors and led to transformations in the supply chain. The COVID-19 pandemic and the requirements to reduce its prevalence have led many people to shop online and encouraged many manufacturers to sell their products online. In this study, a manufacturer, who intends to possess an online sales channel, and a retailer, who has an in-person sales channel, are considered. Then, pricing strategies and collaboration mechanisms between them in the health-social dual-channel supply chain are investigated. This study is developed in three models, including centralized, decentralized, and collaborated under Stackelberg game, whereas the optimal price of products in each channel, level of implementation of health and safety protocols in retailers, advertising level, and status of online shopping performance are obtained for improving customer trust. Moreover, the demand is represented as a function of selling prices of products in online and in-person shops, compliance level of health protocols, level of online shopping performance, and advertising in health during the COVID-19 pandemic. Although the centralized model provides more profit for the manufacturer, the collaborated model provides the highest profit for the retailer. Thus, since the supply chain profit of centralized and collaborated models is close, the collaboration model is the best option for members in this situation. Sensitivity analysis is finally performed to evaluate the impact of key parameters, and then according to obtained results, some management insights are suggested for the dual-channel supply chain during the COVID-19 pandemic.

Hospital dual-channel adoption decisions with telemedicine referral and misdiagnosis.

With the rapid development of telemedicine and the impact of the COVID-19 pandemic, more and more patients are now resorting to using telemedicine channels for healthcare services. However, for hospitals, there exists a lack of managerial guidance in place to help them adopt telemedicine in a practical and standardized way. This study considers a hospital operating with both telemedicine (virtual) and face-to-face (physical) medical channels, and which allocates its capacity by also taking into account the possibility of both referrals and misdiagnosis. Methodologically, we construct a game model based on a queuing framework. We first analyze equilibrium strategies for patient arrivals. Then we propose the necessary conditions for a hospital to develop a telemedicine channel and to operate both channels simultaneously. Finally, we find the optimal decisions for the service level of telemedicine, which can also be regarded as the optimal proportion of diseases treated by telemedicine, and the optimal hospital capacity allocation ratio between the two channels. We also find that hospitals in a full coverage market (e.g., for certain small-scale hospitals and community hospitals or cancer hospitals) are more difficult to adopt telemedicine than hospitals in a partial coverage market (e.g., for comprehensive large-scale hospitals with many potential patients). Small-scale hospitals are more suited to operating telemedicine as a gatekeeper to help triage patients, while large hospitals are more prone to regard telemedicine as a medical channel for providing professional medical services to patients. We also analyze the effects of the telemedicine cure rate and the cost ratio of telemedicine to the physical hospital on the overall healthcare system performance, including the physical hospital arrival rate, patients' waiting time, total profit, and social welfare. Then we compare the performance, ex ante versus ex post, the implementation of telemedicine. It is shown that when the market is partially covered, the total social welfare is always higher than it was before the implementation. However, as far as the profit goes, if the telemedicine cure rate is low and the cost ratio is high, the total hospital profit may be lower than it was prior to using telemedicine. However, the profit and social welfare of hospitals in the full coverage market are always lower than it was before the implementation. In addition, the waiting time in the hospital is always higher than that before the implementation, which means that the implementation of telemedicine will make patients who must receive treatment in the physical hospital face even worse congestion than before. More insights and results are gleaned from a series of numerical studies.

Experimental and numerical evaluation of a novel dual-channel windcatcher with a rotary scoop for energy-saving technology integration

With the increasing requirements for fresh air supply in buildings after the COVID-19 pandemic and the rising energy demand from buildings, there has been an increased emphasis on passive cooling techniques such as natural ventilation. While natural ventilation devices such as windcatchers can be a sustainable and low-cost solution to remove indoor pollutants and improve indoor air quality, it is not as reliable as mechanical systems. Integration with low-energy cooling, heating or heat recovery technologies is necessary for operation in unfavourable outdoor conditions. In this research, a novel dual-channel windcatcher design consisting of a rotary wind scoop and a chimney was proposed to provide a fresh air supply irrespective of the wind direction. The dual-channel design allows for passive cooling, dehumidification and heat recovery technology integration to enhance its thermal performance. In this design, the positions of the supply and return duct are "fixed” or would not change under changing wind directions. An open wind tunnel and test room were employed to experimentally evaluate the ventilation performance of the proposed windcatcher prototype. A Computational Fluid Dynamic (CFD) model was developed and validated to further evaluate the system's ventilation performance. The results confirmed that the system could supply sufficient fresh air and exhaust stale air under changing wind directions. The ventilation rate of the rotary scoop windcatcher was higher than that of a conventional 8-sided multidirectional windcatcher of the same size. © 2023 The Author(s)

Linguistic Pattern-Infused Dual-Channel Bidirectional Long Short-term Memory With Attention for Dengue Case Summary Generation From the Program for Monitoring Emerging Diseases-Mail Database: Algorithm Development Study.

BACKGROUND: Globalization and environmental changes have intensified the emergence or re-emergence of infectious diseases worldwide, such as outbreaks of dengue fever in Southeast Asia. Collaboration on region-wide infectious disease surveillance systems is therefore critical but difficult to achieve because of the different transparency levels of health information systems in different countries. Although the Program for Monitoring Emerging Diseases (ProMED)-mail is the most comprehensive international expert-curated platform providing rich disease outbreak information on humans, animals, and plants, the unstructured text content of the reports makes analysis for further application difficult. OBJECTIVE: To make monitoring the epidemic situation in Southeast Asia more efficient, this study aims to develop an automatic summary of the alert articles from ProMED-mail, a huge textual data source. In this paper, we proposed a text summarization method that uses natural language processing technology to automatically extract important sentences from alert articles in ProMED-mail emails to generate summaries. Using our method, we can quickly capture crucial information to help make important decisions regarding epidemic surveillance. METHODS: Our data, which span a period from 1994 to 2019, come from the ProMED-mail website. We analyzed the collected data to establish a unique Taiwan dengue corpus that was validated with professionals' annotations to achieve almost perfect agreement (Cohen κ=90%). To generate a ProMED-mail summary, we developed a dual-channel bidirectional long short-term memory with attention mechanism with infused latent syntactic features to identify key sentences from the alerting article. RESULTS: Our method is superior to many well-known machine learning and neural network approaches in identifying important sentences, achieving a macroaverage F1 score of 93%. Moreover, it can successfully extract the relevant correct information on dengue fever from a ProMED-mail alerting article, which can help researchers or general users to quickly understand the essence of the alerting article at first glance. In addition to verifying the model, we also recruited 3 professional experts and 2 students from related fields to participate in a satisfaction survey on the generated summaries, and the results show that 84% (63/75) of the summaries received high satisfaction ratings. CONCLUSIONS: The proposed approach successfully fuses latent syntactic features into a deep neural network to analyze the syntactic, semantic, and contextual information in the text. It then exploits the derived information to identify crucial sentences in the ProMED-mail alerting article. The experiment results show that the proposed method is not only effective but also outperforms the compared methods. Our approach also demonstrates the potential for case summary generation from ProMED-mail alerting articles. In terms of practical application, when a new alerting article arrives, our method can quickly identify the relevant case information, which is the most critical part, to use as a reference or for further analysis.

Rapid and universal detection of SARS-CoV-2 and influenza A virus using a reusable dual-channel optic fiber immunosensor.

Establishment of rapid on-site detection technology capable of concurrently detecting SARS-Cov-2 and influenza A virus is urgent to effectively control the epidemic from these two types of important viruses. Accordingly, we developed a reusable dual-channel optical fiber immunosensor (DOFIS), which utilized the evanescent wave-sensing properties and tandem detection mode of the mobile phase, effectively accelerating the detection process such that it can be completed within 10 min. It could detect the nucleoprotein of multiple influenza A viruses (H1N1, H3N2, and H7N9), as well as the spike proteins of the SARS-CoV-2 Omicron and Delta variants, and could respond to 20 TCID50 /ml SARS-CoV-2 pseudovirus and 100 TCID50 /ml influenza A (A/PR/8/H1N1), presenting lower limit of detection and wider linear range than enzyme-linked immunosorbent assay. The detection results on 26 clinical samples for SARS-CoV-2 demonstrated its specificity (100%) and sensitivity (94%), much higher than the sensitivity of commercial colloidal gold test strip (35%). Particularly, DOFIS might be reused more than 80 times, showing not only cost-saving but also potential in real-time monitoring of the pathogenic viruses. Therefore, this newly-developed DOFIS platform is low cost, simple to operate, and has broad spectrum detection capabilities for SARS-CoV-2 mutations and multiple influenza A strains. It may prove suitable for deployment as a rapid on-site screening and surveillance technique for infectious disease.

Achieving Resilience: Resilient Price and Quality Strategies of Fresh Food Dual-Channel Supply Chain Considering the Disruption

In the face of demand disruptions, dual-channel supply chains (SCs) that lack resilience may be more vulnerable. Reaching moderate SC resilience through coordination is essential for dealing with disruptions. This paper investigates the operation management of a dual-channel fresh-food SC (FSC) under disruption. The centralized and decentralized decision models propose joint quality efforts based on the consideration of quality preference and loss. From the perspective of SC resilience, we analyze how SC members can optimally make price, quality, and quantity decisions resiliently and robustly under the disruption of quality preference. The results show that (1) no matter the kind of decision model, considering quality preference disruptions can significantly increase the SC profit;(2) there is a resilience range in decisions with the influence of the disruption cost. The original optimal decisions in the resilience range are robust and sustain SC performance without change;and (3) the disruption significantly impacts offline channel retailers, who are at a disadvantage when competing with online channels. A centralized decision model can achieve higher profits and quality levels in response to demand disruptions. This paper extends the concept of resilience to the FSC and provides suggestions for fresh-food enterprises to conduct quality efforts and cope with demand interruption.

The Impact of CSR on the Performance of a Dual-Channel Closed-Loop Supply Chain under Two Carbon Regulatory Policies

Under different carbon regulatory policies, corporate social responsibility (CSR) activities will have different impacts on the environmental benefits of the supply chain and corporate carbon emission reduction decisions. In this study, we examine a dual-channel closed-loop supply chain consisting of a single manufacturer selling re-products generated from waste products and a single retailer selling new products and consider two settings: enforcing a carbon tax policy or enforcing a subsidy policy. Under each setting, we put CSR into account, construct two models for the retailer to implement or not implement CSR activities, and analyze the decisions obtained under optimal solutions. Through numerical simulation and comparative research, we observe that the carbon tax policy applies to the supply chain where CSR activities are implemented, while the subsidy policy applies to the supply chain where CSR activities are not implemented. Reasonable selection of CSR implementation methods with low-cost coefficients by the retailer is conducive to eliminating profit conflicts among supply chain members. The government should fully consider the decision-making thresholds of supply chain members to ensure the maximum effectiveness of the policy.

A sustainable dual-channel inventory model with trapezoidal fuzzy demand and energy consumption

Most of the consumers relied heavily on e-commerce for products and services for the past few years due to the recent COVID-19 pandemic. This kind of an unexpected behaviour among the consumer society has taken every industry by surprise so that many industries have begun operating online and offline businesses to ensure future competitiveness. Firms introducing online sales are deplorably facing many challenges in terms of logistics and delivery processes, such as short lead times, flexible delivery, capacity of warehouse, and the production process for controlling carbon emissions. Keeping these challenges in mind, a sustainable dual-channel vendor-buyer supply chain model has considered for a controllable emission under fuzzy demand and energy consumption. The model deals with limitation on warehouse floor-space area, and the warehouse divides into two stages such as one for satisfying online orders and the other for satisfying offline orders. The demand rate and energy consumptions are treated as the trapezoidal fuzzy number, and we use the signed distance method to defuzzify the fuzzy joint expected total cost. The objective focuses on obtaining a trade-off between cost and emissions, thereby determining the optimal production-distribution strategy and a proper sustainable plan for handling both online and offline orders. The aforesaid scenario is mathematically formulated in the form of constrained non-linear programme (NLP) and derive a Lagrangean multiplier method to solve it. An iterative solution algorithm is designed, and for better illustration of the developed theory, numerical analysis is carried out followed by a wide discussion on the sensitivity analysis for various parameters. Our results indicate that the optimal solutions of the sustainable fuzzy model slightly fluctuate from the solutions of the sustainable crisp model. According to results, considering the uncertainty in the system is a crucial factor to achieve the economic and environmental sustainability of the production sector. The research reveals that the practitioners should be careful in accounting flexibility in the input factors demand and energy to tackle the uncertainties that always fit the real situation.