Zhejiang Gongshang University

Dysphagia is estimated to affect ~8% of the world's population (~590 million people). Texture-modified foods and thickened drinks are commonly used to reduce the risks of choking and aspiration. The International Dysphagia Diet Standardisation Initiative (IDDSI) was founded with the goal of developing globally standardized terminology and definitions for texture-modified foods and liquids applicable to individuals with dysphagia of all ages, in all care settings, and all cultures. A multi-professional volunteer committee developed a dysphagia diet framework through systematic review and stakeholder consultation. First, a survey of existing national terminologies and current practice was conducted, receiving 2050 responses from 33 countries. Respondents included individuals with dysphagia; their caregivers; organizations supporting individuals with dysphagia; healthcare professionals; food service providers; researchers; and industry. The results revealed common use of 3-4 levels of food texture (54 different names) and ≥3 levels of liquid thickness (27 different names). Substantial support was expressed for international standardization. Next, a systematic review regarding the impact of food texture and liquid consistency on swallowing was completed. A meeting was then convened to review data from previous phases, and develop a draft framework. A further international stakeholder survey sought feedback to guide framework refinement; 3190 responses were received from 57 countries. The IDDSI Framework (released in November, 2015) involves a continuum of 8 levels (0-7) identified by numbers, text labels, color codes, definitions, and measurement methods. The IDDSI Framework is recommended for implementation throughout the world.

Highly photoactive MOFs can be engineered <italic>via</italic> various strategies for the purpose of extended visible light absorption, more efficient generation, separation and transfer of charge carriers, as well as good recyclability.

The effectiveness of environmental, social, and corporate governance (ESG) ratings has been widely discussed and is often linked to corporate financial performance by academics and practitioners. However, a significant research gap remains unexplored, specifically, measuring the underlying mechanisms between ESG ratings and corporate green innovation in developing countries. This study unpacks the quasi-natural experiment based on the 2015 ESG rating of the SynTao Green Finance Agency to investigate how ESG ratings affect corporate green innovation based on data relating to Chinese A-share listed companies between 2010 and 2018. The results show that ESG ratings significantly promote the quantity and quality of corporate green innovation and are mediated by alleviating financial constraints and increasing managers' environmental awareness. In this context, the higher the ESG rating score, the more apparent is the promotion effect. Furthermore, stricter environmental regulations, increased market competition, and companies in their growth period strengthen the association between ESG ratings and green innovation. This study provides scientific evidence for the role of ESG ratings in promoting proactive green innovation and deepening green development in China.

Texture modification has become one of the most common forms of intervention for dysphagia, and is widely considered important for promoting safe and efficient swallowing. However, to date, there is no single convention with respect to the terminology used to describe levels of liquid thickening or food texture modification for clinical use. As a first step toward building a common taxonomy, a systematic review was undertaken to identify empirical evidence describing the impact of liquid consistency and food texture on swallowing behavior. A multi-engine search yielded 10,147 non-duplicate articles, which were screened for relevance. A team of ten international researchers collaborated to conduct full-text reviews for 488 of these articles, which met the study inclusion criteria. Of these, 36 articles were found to contain specific information comparing oral processing or swallowing behaviors for at least two liquid consistencies or food textures. Qualitative synthesis revealed two key trends with respect to the impact of thickening liquids on swallowing: thicker liquids reduce the risk of penetration-aspiration, but also increase the risk of post-swallow residue in the pharynx. The literature was insufficient to support the delineation of specific viscosity boundaries or other quantifiable material properties related to these clinical outcomes. With respect to food texture, the literature pointed to properties of hardness, cohesiveness, and slipperiness as being relevant both for physiological behaviors and bolus flow patterns. The literature suggests a need to classify food and fluid behavior in the context of the physiological processes involved in oral transport and flow initiation.

Knowledge graph (KG) plays an increasingly important role in recommender systems. A recent technical trend is to develop end-to-end models founded on graph neural networks (GNNs). However, existing GNN-based models are coarse-grained in relational modeling, failing to (1) identify user-item relation at a fine-grained level of intents, and (2) exploit relation dependencies to preserve the semantics of long-range connectivity.

The outbreak of Corona Virus Disease 2019 (COVID-19) in various countries at the end of last year has transferred traditional face-to-face teaching to online education platforms, which directly affects the quality of education. Taking user satisfaction on online education platforms in China as the research object, this paper uses a questionnaire survey and web crawler to collect experience data of online and offline users, constructs a customer satisfaction index system by analyzing emotion and the existing literature for quantitative analysis, and builds aback propagation (BP) neural network model to forecast user satisfaction. The conclusion shows that users' personal factors have no direct influence on user satisfaction, while platform availability has the greatest influence on user satisfaction. Finally, suggestions on improving the online education platform are given to escalate the level of online education during the COVID-19 pandemic, so as to promote the reform of information-based education.

Uncovering the complex interactions among ecosystem services (ESs) is a prerequisite for managing multiple ESs simultaneously. We mapped the supply of six ESs, quantified their trade-offs/synergies and bundles, determined dominant social-ecological drivers, and subsequently provided sustainable spatial planning and management strategies at both grid and sub-watershed scales. The results revealed that: (1) the patterns of ESs were spatially heterogeneous, with food production decreasing the most, by 14.86% (grid scale) and 17% (sub-watershed scale), and water supply increasing the most, by 22.46% (grid scale) and 22.57% (sub-watershed scale) from 2000 to 2020; (2) ten ES pairs presented synergies while five ES pairs related to food production showed trade-offs at both spatial scales, with three ES pairs at the grid scale and two ES pairs at the sub-watershed scale experiencing declined synergies or increased trade-offs; (3) our spatial-temporal and cross-scale analysis of ES interactions identified ES pairs with declined synergies as an additional priority in ecosystem management and determined ES bundles as a zoning basis for spatial planning at both scales. The similarities and divergences in ES interactions and their drivers at different scales require the linkage and distinct focus of spatial planning at different scales. Thus, we integrated spatial-temporal and cross-scale knowledge on ES interactions into spatial planning to underpin sustainable ecosystem management across scales.

Plants play a crucial role in supplying food globally. Various environmental factors lead to plant diseases which results in significant production losses. However, manual detection of plant diseases is a time-consuming and error-prone process. It can be an unreliable method of identifying and preventing the spread of plant diseases. Adopting advanced technologies such as Machine Learning (ML) and Deep Learning (DL) can help to overcome these challenges by enabling early identification of plant diseases. In this paper, the recent advancements in the use of ML and DL techniques for the identification of plant diseases are explored. The research focuses on publications between 2015 and 2022, and the experiments discussed in this study demonstrate the effectiveness of using these techniques in improving the accuracy and efficiency of plant disease detection. This study also addresses the challenges and limitations associated with using ML and DL for plant disease identification, such as issues with data availability, imaging quality, and the differentiation between healthy and diseased plants. The research provides valuable insights for plant disease detection researchers, practitioners, and industry professionals by offering solutions to these challenges and limitations, providing a comprehensive understanding of the current state of research in this field, highlighting the benefits and limitations of these methods, and proposing potential solutions to overcome the challenges of their implementation.

Nearest neighbor search is a fundamental and essential operation in applications from many domains, such as databases, machine learning, multimedia, and computer vision. Because exact searching results are not efficient for a high-dimensional space, a lot of efforts have turned to approximate nearest neighbor search. Although many algorithms have been continuously proposed in the literature each year, there is no comprehensive evaluation and analysis of their performance. In this paper, we conduct a comprehensive experimental evaluation of many state-of-the-art methods for approximate nearest neighbor search. Our study (1) is cross-disciplinary (i.e., including 19 algorithms in different domains, and from practitioners) and (2) has evaluated a diverse range of settings, including 20 datasets, several evaluation metrics, and different query workloads. The experimental results are carefully reported and analyzed to understand the performance results. Furthermore, we propose a new method that achieves both high query efficiency and high recall empirically on majority of the datasets under a wide range of settings.

Deep learning architecture with convolutional neural network achieves outstanding success in the field of computer vision. Where U-Net has made a great breakthrough in biomedical image segmentation and has been widely applied in a wide range of practical scenarios. However, the equal design of every downsampling layer in the encoder part and simply stacked convolutions do not allow U-Net to extract sufficient information of features from different depths. The increasing complexity of medical images brings new challenges to the existing methods. In this paper, we propose a deeper and more compact split-attention u-shape network, which efficiently utilises low-level and high-level semantic information based on two frameworks: primary feature conservation and compact split-attention block. We evaluate the proposed model on CVC-ClinicDB, 2018 Data Science Bowl, ISIC-2018, SegPC-2021 and BraTS-2021 datasets. As a result, our proposed model displays better performance than other state-of-the-art methods in terms of the mean intersection over union and dice coefficient. More significantly, the proposed model demonstrates excellent segmentation performance on challenging images. The code for our work and more technical details can be found at https://github.com/xq141839/DCSAU-Net.

Distributed Denial of Service (DDoS) attacks grow rapidly and become one of the fatal threats to the Internet. Automatically detecting DDoS attack packets is one of the main defense mechanisms. Conventional solutions monitor network traffic and identify attack activities from legitimate network traffic based on statistical divergence. Machine learning is another method to improve identifying performance based on statistical features. However, conventional machine learning techniques are limited by the shallow representation models. In this paper, we propose a deep learning based DDoS attack detection approach (DeepDefense). Deep learning approach can automatically extract high-level features from low-level ones and gain powerful representation and inference. We design a recurrent deep neural network to learn patterns from sequences of network traffic and trace network attack activities. The experimental results demonstrate a better performance of our model compared with conventional machine learning models. We reduce the error rate from 7.517% to 2.103% compared with conventional machine learning method in the larger data set.

The security problems of smart contracts have drawn extensive attention due to the enormous financial losses caused by vulnerabilities. Existing methods on smart contract vulnerability detection heavily rely on fixed expert rules, leading to low detection accuracy. In this paper, we explore using graph neural networks (GNNs) for smart contract vulnerability detection. Particularly, we construct a contract graph to represent both syntactic and semantic structures of a smart contract function. To highlight the major nodes, we design an elimination phase to normalize the graph. Then, we propose a degree-free graph convolutional neural network (DR-GCN) and a novel temporal message propagation network (TMP) to learn from the normalized graphs for vulnerability detection. Extensive experiments show that our proposed approach significantly outperforms state-of-the-art methods in detecting three different types of vulnerabilities.

Based on the existing research on environmental regulation (ER) and enterprise innovation, this paper separates green innovation ability from general technological innovation and divides it, from the perspective of motivation, into substantive green innovation (SUBGI) and symbolic green innovation (SYMGI). This paper employs panel data of China's A-share listed enterprise from 2008 to 2018 in a difference-in-difference-in-difference model to construct a quasinatural experiment on the impacts of ER and the green innovation strategy of enterprises. Government subsidies and regulatory capture are used to explore the mechanism between ER and innovation behavior. The results show that ER has a significantly positive effect on green innovation, but its impact on SYMGI and SUBGI decreases. Under the constraints of the ER policy, government subsidies incentivize the green innovation of enterprises, but they are not the main reason for the difference between the two innovation behaviors. Regulatory capture plays a negative moderating role in ER promoting enterprises’ SUBGI and has no significant impact on SYMGI, which is the key factor leading to the difference between the two innovation behaviors. The heterogeneity test suggests a pronounced promotion effect of ER on state-owned enterprises, large enterprises, and growing enterprises. This study provides momentous policy implications for making rational use of environmental policies in promoting enterprises’ green innovation capability, especially high-quality green innovation behavior aimed at promoting enterprises’ green technology progress and gaining competitive advantages.

In this paper, we propose a new unsupervised spectral feature selection model by embedding a graph regularizer into the framework of joint sparse regression for preserving the local structures of data. To do this, we first extract the bases of training data by previous dictionary learning methods and, then, map original data into the basis space to generate their new representations, by proposing a novel joint graph sparse coding (JGSC) model. In JGSC, we first formulate its objective function by simultaneously taking subspace learning and joint sparse regression into account, then, design a new optimization solution to solve the resulting objective function, and further prove the convergence of the proposed solution. Furthermore, we extend JGSC to a robust JGSC (RJGSC) via replacing the least square loss function with a robust loss function, for achieving the same goals and also avoiding the impact of outliers. Finally, experimental results on real data sets showed that both JGSC and RJGSC outperformed the state-of-the-art algorithms in terms of k -nearest neighbor classification performance.

The rapid urbanization witnessed in developing countries in Asia and Africa has led to a substantial increase in municipal solid waste (MSW) generation. However, the corresponding disposal strategies, along with constraints in land resources and finances, compounded by unorganized public behaviour, have resulted in ineffective policy implementation and monitoring. This lack of systematic and targeted orientation, combined with blind mapping, has led to inefficient development in many areas. This review examines the key challenges of MSW management in developing countries in Asia and Africa from 2013 to 2023, drawing insights from 170 academic papers. Rather than solely focusing on recycling, the study proposes waste sorting at the source, optimization of landfill practices, thermal treatment measures, and strategies to capitalize on the value of waste as more pertinent solutions aligned with local realities. Barriers to optimizing management systems arise from socio-economic factors, infrastructural limitations, and cultural considerations. The review emphasizes the importance of integrating the study area into the circular economy framework, with a focus on enhancing citizen participation in solid waste reduction and promoting recycling initiatives, along with seeking economic assistance from international organizations.

Curcumin is a bioactive constituent isolated from turmeric that has historically been used as a seasoning, pigment, and herbal medicine in food. Recently, it has become one of the most commonly studied nutraceuticals in the pharmaceutical, supplement, and food areas because of its myriad of potential health benefits. For instance, it is claimed to exhibit antioxidant, anti-inflammatory, antimicrobial, antiparasite, and anticancer activities when ingested as a drug, supplement, or food. Toxicity studies suggest that it is safe to consume, even at relatively high levels. Its broad-spectrum biological activities and low toxicity have meant that it has been widely explored as a nutraceutical ingredient for application in functional foods. However, there are several hurdles that formulators must overcome when incorporating curcumin into commercial products, such as its low water solubility (especially under acidic and neutral conditions), chemical instability (especially under neutral and alkaline conditions), rapid metabolism by enzymes in the human body, and limited bioavailability. As a result, only a small fraction of ingested curcumin is actually absorbed into the bloodstream. These hurdles can be at least partially overcome by using encapsulation technologies, which involve trapping the curcumin within small particles. Some of the most commonly used edible microparticles or nanoparticles utilized for this purpose are micelles, liposomes, emulsions, solid lipid particles, and biopolymer particles. Each of these encapsulation technologies has its own benefits and limitations for particular product applications and it is important to select the most appropriate one.

Financial technology, or FinTech, involves the design and delivery of financial products and services through technology. It impacts financial institutions, regulators, customers, and merchants across a wide range of industries. Pervasive digital technologies are challenging the fundamentals of the highly regulated financial sector, leading to the emergence of non-traditional payment systems, peer-to-peer money exchanges and increased turbulence in currency markets. This case study explores the development of a FinTech company in China that offers microloans to college students. Five lessons learned are presented for organizations to better manage the challenges and to leverage the opportunities amidst the disruption of financial sector. Our findings also shed light on how digital technology 1) offers the strategic capability for a firm to occupy a market niche in financial sector, 2) enables the generation of alternative credit scores based on non-traditional data, and 3) improves the financial inclusion of previously excluded market segments.

Big data, because it can mine new knowledge for economic growth and technical innovation, has recently received considerable attention, and many research efforts have been directed to big data processing due to its high volume, velocity, and variety (referred to as "3V") challenges. However, in addition to the 3V challenges, the flourishing of big data also hinges on fully understanding and managing newly arising security and privacy challenges. If data are not authentic, new mined knowledge will be unconvincing; while if privacy is not well addressed, people may be reluctant to share their data. Because security has been investigated as a new dimension, "veracity," in big data, in this article, we aim to exploit new challenges of big data in terms of privacy, and devote our attention toward efficient and privacy-preserving computing in the big data era. Specifically, we first formalize the general architecture of big data analytics, identify the corresponding privacy requirements, and introduce an efficient and privacy-preserving cosine similarity computing protocol as an example in response to data mining's efficiency and privacy requirements in the big data era.

In image analysis, the images are often represented by multiple visual features (also known as multiview features), that aim to better interpret them for achieving remarkable performance of the learning. Since the processes of feature extraction on each view are separated, the multiple visual features of images may include overlap, noise, and redundancy. Thus, learning with all the derived views of the data could decrease the effectiveness. To address this, this paper simultaneously conducts a hierarchical feature selection and a multiview multilabel (MVML) learning for multiview image classification, via embedding a proposed a new block-row regularizer into the MVML framework. The block-row regularizer concatenating a Frobenius norm (F-norm) regularizer and an l(2,1)-norm regularizer is designed to conduct a hierarchical feature selection, in which the F-norm regularizer is used to conduct a high-level feature selection for selecting the informative views (i.e., discarding the uninformative views) and the l(2,1)-norm regularizer is then used to conduct a low-level feature selection on the informative views. The rationale of the use of a block-row regularizer is to avoid the issue of the over-fitting (via the block-row regularizer), to remove redundant views and to preserve the natural group structures of data (via the F-norm regularizer), and to remove noisy features (the l(2,1)-norm regularizer), respectively. We further devise a computationally efficient algorithm to optimize the derived objective function and also theoretically prove the convergence of the proposed optimization method. Finally, the results on real image datasets show that the proposed method outperforms two baseline algorithms and three state-of-the-art algorithms in terms of classification performance.

Dietary proteins and phenolic compounds are commonly co-existing components that readily interact with each other to yield complexes in a wide range of food systems. The formed complexes play a critical role in the physiochemical characteristics of both reacting molecules, thereby impacting nutritional and quality profiles of related products. In this review, we provided the most updated knowledge on dietary protein-phenolic interactions related with food science and human nutrition, including their mechanisms of complexation, analytical technologies, and alterations in the functionality and nutraceutical properties of both reacting partners. Their potential applications in the industries regarding stability during food processing and storage, impacts on product quality, and fabrication of novel delivery systems for liable bioactives were also discussed. The interactions between dietary proteins and phenolics, either via non-covalent or covalent processes, are ubiquitous in food systems and are closely associated with chemical structures of both compounds and the surrounding conditions, mainly temperature, pH, and the presence of phenolic oxidases. Albeit in different ways, such intermolecular associations induced changes in protein conformational structures, which subsequently impacted their techno-functional properties, digestibility, and allergenic potentials; in turn, the bioaccessibility/bioavailability and health-protecting features of interacted phenolics were modified to various extents, as noticed by in vitro and in vivo evidence. Largely depending on the interaction molecules and preparation steps, those influences can be either favorable or unfavorable in different systems and therefore can be tailored to develop food products and nutraceuticals with maximized functionality and quality attributes.