Search

Despite its low energy intensity, the shipping industry accounts for roughly 3% of total anthropogenic Greenhouse Gas (GHG) emissions worldwide according to IMO data for 2020, and concerns regarding the environmental effects of marine transportation have grown over recent decades. Major countries have stepped up initiatives to tackle this challenge. This study explores the evolution of China's green shipping policies and regulations, with particular attention to the collaborative relationships among policy entities at different levels, their respective roles, and the changing trends in policy thematic emphasis. Over the period from 2009 to June 2022, 208 green policy documents were manually collected from the official websites of various policy entities. These documents underwent analysis using a social network analysis methodology, with variables such as density, degree centrality, betweenness centrality, n-clan, and k-core being examined. Results indicate that China's green shipping regulations have undergone significant transitions in recent years. The focus of policies has shifted from mitigating atmospheric pollutant emissions to reducing GHG emissions. Notable, there has been a greater emphasis on promoting green energy rather than solely improving energy efficiency. Governments seem increasingly inclined to implement mandatory green shipping policies over voluntary ones to ensure their practical effectiveness. Furthermore, recent decentralized administrative reforms have empowered local governments with greater decision-making authority. Gaining insight into the evolution and changing thematic priorities of China's green shipping regulations is essential for policymakers and researchers to develop effective strategies for addressing environmental concerns within the maritime sector. This understanding also aids industry practitioners in ensuring compliance with regulatory standards and fostering sustainable shipping practices. • Policy actors work together extensively to implement green shipping regulations. • China's green shipping policies exhibit four main traits. • Green shipping policy network is characterized by the "pyramid-like" structure. [ABSTRACT FROM AUTHOR]

Biomass chemical looping gasification (BCLG) is an emerging technology that enables waste co-processing into clean energy. A laboratory-scale vertical fixed-bed reactor was used to conduct CLG experiments with vinegar residue (VR) as a biomass fuel, and the effect of enhanced LD slag composite Ca-Fe material as an oxygen carrier (OC) on the production of syngas from BCLG was investigated. This study analyzed the influence of LD slag proportion, kinds of OC, and operational conditions (temperature, oxygen carrier/biomass (OC/B) ratio, and steam/biomass (S/B) ratio) on gasification performance. Results indicate that the syngas yield under the gasification operating conditions at about 850 °C, OC/B = 1, and S/B = 2.5 obtained 1.36 Nm3/kg, the CCE and CGE achieved about 80.82 % and 98.16 %, respectively, and the H 2 /CO molar ratio was as high as 2.57. The OCs multi-redox cycle test revealed that after undergoing 10 redox cycles under the conditions mentioned above, the gas yield, CCE, and CGE were reduced to varying extents, but still maintained impressive reactivity. The mechanism analysis showed that the performance degradation of the LD slag composite Ca-Fe OC during the multi-cycle process was related to its agglomeration due to thermal stress and steam. This study provides a new perspective on the synthesis of low-cost OCs and the co-utilization of solid wastes. [Display omitted] • Enhanced LD slag Ca-Fe composite oxygen carrier was prepared and used for biomass chemical looping gasification. • The operating conditions were screened and optimized to obtain H 2 /CO syngas with a regulated ratio. • The reaction mechanism for producing syngas from waste co-utilization chemical looping gasification is summarized. [ABSTRACT FROM AUTHOR]

Rehmannia piasezkii is a kind of medicinal plants, of the Orobanchaceae family, and well known for its large pink or purple corolla. However, no research on the molecular mechanism of flower color formation in R. piasezkii has been conducted so far. In this study, we investigated the transcriptome of root, stem, leaf and corollas of R. piasezkii using transcriptome sequencing technology and assembled 144,582 unigenes. A total of 58 anthocyanin biosynthetic genes were identified in the R. piasezkii transcriptome, fourteen of which were highly correlated with anthocyanin content, especially RpF3H2 , RpDFR2 , RpANS1 , RpANS2 and RpUFGT. Totally, 35 MYB genes with FPKM values greater than 5 were identified in the R. piasezkii transcriptome, including an R2R3 MYB transcriptional factor RpMYB1 , which belongs to subgroup 6 of the R2R3 MYB family. Agrobacterium -mediated transient expression of Nicotiana benthamiana revealed that overexpression of RpMYB1 could activate the expression of structural genes in anthocyanin synthesis pathway and promote the accumulation of anthocyanins in N. benthamiana leaves, indicating that RpMYB1 is a positive regulator of anthocyanin synthesis. Furthermore, combined transient overexpression of RpMYB1 with RpANS1 , RpMYB1 + RpANS1 with other structural genes all could further enhance the accumulation of anthocyanins in N. benthamiana leaves. Permanent overexpression of RpMYB1 in R. glutinosa promoted anthocyanin accumulation and expression levels of RgCHS , RgF3H , RgDFR and RgANS. Further evidence from dual-luciferase assay suggested that RpMYB1 could bind to the promoter of RpDFR2 and hence activating its expression. These findings provide insight into the molecular regulation in anthocyanin biosynthesis in R. piasezkii and provide valuable genetic resources for the genetic improvement of flower color. • The transcriptome sequencing of R. piasezkii was carried out for the first time in the world. • A total of 58 anthocyanin biosynthetic genes were identified in R. piasezkii transcriptome. • RpMYB1 is the key transcriptional activator of anthocyanin biosynthesis in R. piasezkii. • Engineering of RpMYB1 in R. glutinosa could provide anthocyanin-rich foods. [ABSTRACT FROM AUTHOR]

Computing geodesic paths and distances is a common operation in computer graphics and computer-aided geometric design. The existing discrete geodesic algorithms are mainly designed to solve the boundary value problem, i.e., to find the shortest path between two given points. In this paper, we focus on the initial value problem, i.e., finding a uniquely determined geodesic path from a given point in any direction. Since the shortest paths do not provide the unique solution on triangle meshes, we solve the initial value problem in an indirect manner: given a fixed point and an initial tangent direction on a triangle mesh M , we first compute a geodesic curve γ ̂ on a piecewise smooth surface M ̂ , which well approximates the input mesh M and can be constructed at little cost. Then, we solve a first-order ODE of the tangent vector using the fourth-order Runge–Kutta method, and parallel transport it along γ ̂ . When the geodesic curve reaches the boundary of the current patch, its tangent can be directly transported to the neighboring patch, thanks to the G 1 -continuity along the common boundary of two adjacent patches. Finally, once the geodesic curve γ ̂ is available, we project it onto the underlying mesh M , producing the discrete geodesic path γ , which is guaranteed to be unique on M . It is worth noting that our method is different from the conventional methods of directly solving the geodesic equation (i.e., a second-order ODE of the position) on piecewise smooth surfaces, which are difficult to implement due to the complicated representation of the geodesic equation involving Christoffel symbols. The proposed method, based on the first-order ODE of the tangent vector, is intuitive and easy for implementation. Our method is particularly useful for computing geodesic paths on low-resolution meshes which may have large and/or skinny triangles, since the conventional straightest geodesic paths are usually far from the ground truth. [ABSTRACT FROM AUTHOR]

Sparse data is known to pose challenges to cluster analysis, as the similarity between data tends to be ill-posed in the high-dimensional Hilbert space. Solutions in the literature typically extend either k-means or spectral clustering with additional steps on representation learning and/or feature weighting. However, adding these usually introduces new parameters and increases computational cost, thus inevitably lowering the robustness of these algorithms when handling massive ill-represented data. To alleviate these issues, this paper presents a class of self-organizing neural networks, called the salience-aware adaptive resonance theory (SA-ART) model. SA-ART extends Fuzzy ART with measures for cluster-wise salient feature modeling. Specifically, two strategies, i.e. cluster space matching and salience feature weighting, are incorporated to alleviate the side-effect of noisy features incurred by high dimensionality. Additionally, cluster weights are bounded by the statistical means and minimums of the samples therein, making the learning rate also self-adaptable. Notably, SA-ART allows clusters to have their own sets of self-adaptable parameters. It has the same time complexity of Fuzzy ART and does not introduce additional hyperparameters that profile cluster properties. Comparative experiments have been conducted on the ImageNet and BlogCatalog datasets, which are large-scale and include sparsely-represented data. The results show that, SA-ART achieves 51.8% and 18.2% improvement over Fuzzy ART, respectively. While both have a similar time cost, SA-ART converges faster and can reach a better local minimum. In addition, SA-ART consistently outperforms six other state-of-the-art algorithms in terms of precision and F1 score. More importantly, it is much faster and exhibits stronger robustness to large and complex data. [ABSTRACT FROM AUTHOR]

Estimated Time of Arrival (ETA) for packages plays an essential role in intelligent logistics. As a classic ETA method, Origin–Destination-based (OD-based) ETA predicts the delivery time only based on the attributes (i.e. , sender address, receiver address, seller, and payment time) of packages under the condition that the delivery route is unavailable. However, existing OD-based methods only exploit attributes associated with an individual order, which fails to model the higher-order interactions within orders and attributes, and fail to sufficiently exploit the graph-structure knowledge (i.e. , relation of orders and attributes) and feature-based knowledge (i.e. , statistical properties) of orders simultaneously, resulting in inaccurate predictions. In this paper, we propose a novel Heterogeneous HyperGraph Neural Network (H 2 GNN) for estimating package arrival time. Specifically, to better capture the high-order interactions within orders and attributes, we construct an order heterogeneous hypergraph that utilizes hyperedges to represent orders and nodes to represent order attributes. Besides, we extend the hypergraph learning for large-scale e-commerce data by Hyper-GraphSAGE. Overall, H 2 GNN can provide informatively representations of packages while preserving both structure-based knowledge learned by hypergraph and feature-based knowledge captured by Transformer. Experimental results on large-scale Alibaba logistics data demonstrate the superior performance of H 2 GNN compared to the baselines. [ABSTRACT FROM AUTHOR]

• We divide the recipes into phases, and use a global structure prediction component to assign different subgenerators to generate recipe phases. • We incorporate the attention mechanism to get the phase-aware features, which are the input for different subgenerators to produce better recipe phases. • Our proposed framework outperforms previous state-of-the-art results. Recipe generation from food images and ingredients is a challenging task, which requires the interpretation of the information from another modality. Different from the image captioning task, where the captions usually have one sentence, cooking instructions contain multiple sentences and have obvious structures. To help the model capture the recipe structure and avoid missing some cooking details, we propose a novel framework: Decomposing Generation Networks (DGN) with structure prediction, to get more structured and complete recipe generation outputs. Specifically, we split each cooking instruction into several phases, and assign different sub-generators to each phase. Our approach includes two novel ideas: (i) learning the recipe structures with the global structure prediction component and (ii) producing recipe phases in the sub-generator output component based on the predicted structure. Extensive experiments on the challenging large-scale Recipe1M dataset validate the effectiveness of our proposed model, which improves the performance over the state-of-the-art results. [ABSTRACT FROM AUTHOR]

The target of product attributes prediction is to complete the characteristics set for defining a particular product. Most of the existing methods treat the product attributes prediction as a Named-Entity Recognition (NER) problem from the products' affiliated data, such as product title and introduction. However, in a large number of industrial applications of Alibaba, we found that the existing methods are good at concrete attributes extraction (e.g., color, size) but short of abstract attributes extraction (e.g., applicable event). Moreover, these abstract attributes are usually not easy to extract from the products' affiliated data. In this paper, we propose a novel heterogeneous s tar graph a ttention n etwork called "SAN", which incorporates the advantages of multiple information in the e-commerce scene to predict the abstract attributes of products. Specifically, we model the customer interactive behaviors, product title and concrete attributes of a product as a star graph. Then, we extract the node features, node types and graph structure information from the heterogeneous star graph network which consists of n star graphs. By leveraging the parallel multiple attention mechanism, SAN can aggregate features and learn weights of nodes for product representation and abstract attribute prediction. Extensive experimental results of a real-world e-commerce dataset have demonstrated that SAN outperforms state-of-the-art methods significantly for product attributes prediction. These series of solutions are already planned for use in the applications of Alibaba. [ABSTRACT FROM AUTHOR]