Plant disease, a huge burden, can cause yield loss of up to 100% and thus reduce food security. Actually, smart diagnosing diseases with plant phenomics is crucial for recovering the most yield loss, which usually requires sufficient image information. Hence, phenomics is being pursued as an independent discipline to enable the development of high-throughput phenotyping for plant disease. However, we often face challenges in sharing large-scale image data due to incompatibilities in formats and descriptions provided by different communities, limiting multidisciplinary research exploration. To this end, we build a Plant Phenomics Analysis of Disease (PlantPAD) platform with large-scale information on disease. Our platform contains 421 314 images, 63 crops and 310 diseases. Compared to other databases, PlantPAD has extensive, well-annotated image data and in-depth disease information, and offers pre-trained deep-learning models for accurate plant disease diagnosis. PlantPAD supports various valuable applications across multiple disciplines, including intelligent disease diagnosis, disease education and efficient disease detection and control. Through three applications of PlantPAD, we show the easy-to-use and convenient functions. PlantPAD is mainly oriented towards biologists, computer scientists, plant pathologists, farm managers and pesticide scientists, which may easily explore multidisciplinary research to fight against plant diseases. PlantPAD is freely available at http://plantpad.samlab.cn.

Knowledge graphs (KGs) play an important role in many real-world applications like information retrieval, question answering, relation extraction, etc. To reveal implicit knowledge from a knowledge graph (KG), viz. knowledge graph completion (KGC), is a crucial task for the downstream applications based on KG. For this purpose various embedding-based approaches have been proposed recently. This paper proposes a new approach named HRESCAL to KGC. It extends the well-known embedding-based approach RESCAL by introducing Hamming distance-based encoder to capture implicit multihop and partial inverse relation features in a KG. Experimental results on widely used KGC benchmarks show that the new approach achieves state-of-the-art or is competitive AUC performance.

Metadata has been analyzed and summarized. Based on Dublin Core metadata, combined with the characteristics and forms of intangible cultural heritage, this article explores the metadata for intangible cultural heritage in knowledge organizations based on relevant resource description standards. The Wuhan woodcarving ship model is presented as an example of national intangible cultural heritage to control the application of metadata in intangible cultural heritage knowledge organizations. New ideas are provided for the digital development of intangible cultural heritage.