Search

Abstract Enhancing the quality and safety of exported agricultural products and improving export competitiveness is the key to establishing enhanced competitive advantages in agricultural products, developing a trade powerhouse and realising high-quality development of agriculture. This paper uses the data of Chinese listed companies and Chinese Customs from 2007 to 2016 to discuss the effect and mechanism of digital transformation of enterprises on the quality and safety level of export agricultural products by using the staggered differential method. The study shows that (1) Enterprise digital transformation effectively improves the quality and safety of exported agricultural products, and this result holds after endogeneity, placebo and multiple robustness tests; (2) Heterogeneity analyses reveal that the quality and safety effect of enterprise digital transformation is greater for exporting to developed countries’ markets, non-state-owned enterprises and enterprises in the eastern region, in addition to bulk agricultural products and consumer-oriented agricultural products; (3) Mechanism analyses shows that enterprise digital transformation raises the quality and safety of exported agricultural products through technological innovation, product tracing, information sharing and quality assurance effects.

Abstract Improving the quality of agricultural exports and increasing value–added trade is crucial for shaping new competitive advantages of agricultural exports, building a strong trading nation and achieving high–quality agricultural development. This study constructed a multi–time point difference–in–differences model to empirically test the effect and mechanism of the mutual recognition of geographical indications (GIs) between China and the European Union (EU) on the quality upgrading of China’s exported agricultural products from 2000 to 2016 based on theoretical analysis of the effect mechanism of GIs on agricultural product exports. The study determined that mutual GI recognition between China and the EU has effectively improved the quality of agricultural exports, with a greater effect on upgrading the quality of products from countries with high GIs endowment, quality frontier and large and medium–sized enterprises and labour–intensive products. The mechanism analysis revealed that mutual GI recognition between China and the EU can improve the quality of agricultural exports through specialisation agglomeration and cost-saving effects on the supply side and domestic demand upgrading and product recognition effects on the demand side. This study has important implications for further enriching quality improvement theory regarding agricultural products and presents a new approach for enhancing the quality improvement path of exported agricultural products that lack supply–side resources.

Human action recognition has facilitated the development of artificial intelligence devices focusing on human activities and services. This technology has progressed by introducing 3D point clouds derived from depth cameras or radars. However, human behavior is intricate, and the involved point clouds are vast, disordered, and complicated, posing challenges to 3D action recognition. To solve these problems, we propose a Symmetric Fine-coarse Neural Network (SFCNet) that simultaneously analyzes human actions’ appearance and details. Firstly, the point cloud sequences are transformed and voxelized into structured 3D voxel sets. These sets are then augmented with an interval-frequency descriptor to generate 6D features capturing spatiotemporal dynamic information. By evaluating voxel space occupancy using thresholding, we can effectively identify the essential parts. After that, all the voxels with the 6D feature are directed to the global coarse stream, while the voxels within the key parts are routed to the local fine stream. These two streams extract global appearance features and critical body parts by utilizing symmetric PointNet++. Subsequently, attention feature fusion is employed to capture more discriminative motion patterns adaptively. Experiments conducted on public benchmark datasets NTU RGB+D 60 and NTU RGB+D 120 validate SFCNet’s effectiveness and superiority for 3D action recognition.

Geographical indication (GI) represents the specific good quality and reputation of the regional characteristics of agricultural products, which is a positive approach for stabilising the export of agricultural products, realising high quality and high prices for agricultural products and rural revitalisation. Based on the China-EU Geographical Indications Agreement, this paper discusses the differences between China and the European Union (EU) in terms of GI protection from three aspects: the institutional framework, the operational system, and the operational status of GIs. The purpose of the study is to identify the differences between China and the EU in the protection of GIs, China's shortcomings, and gaps, and to propose policy recommendations for China to protect GI products better, exploit the economic impact of GIs and enhance the competitiveness of China's exports.

Social-network-based recommendation algorithms leverage rich social network information to alleviate the problem of data sparsity and boost the recommendation performance. However, traditional social-network-based recommendation algorithms ignore high-order collaborative signals or only consider the first-order collaborative signal when learning users’ and items’ latent representations, resulting in suboptimal recommendation performance. In this paper, we propose a graph neural network (GNN)-based social recommendation model that utilizes the GNN framework to capture high-order collaborative signals in the process of learning the latent representations of users and items. Specifically, we formulate the representations of entities, i.e., users and items, by stacking multiple embedding propagation layers to recursively aggregate multi-hop neighborhood information on both the user–item interaction graph and the social network graph. Hence, the collaborative signals hidden in both the user–item interaction graph and the social network graph are explicitly injected into the final representations of entities. Moreover, we ease the training process of the proposed GNN-based social recommendation model and alleviate overfitting by adopting a lightweight GNN framework that only retains the neighborhood aggregation component and abandons the feature transformation and nonlinear activation components. The experimental results on two real-world datasets show that our proposed GNN-based social recommendation method outperforms the state-of-the-art recommendation algorithms.