[Purpose/Significance] Advances in single-cell sequencing and high-throughput technology have made it possible for plant genomics to accumulate large quantities of data describing multidimensional genomic-wide molecular phenotypes at low cost. As powerful data mining tools, deep learning techniques can be utilized to further predict and interpret the acquired molecular phenotypes. In recent studies, deep learning has been shown to yield significant results in plant genomics and crop breeding research. However, a complete review of deep learning applications in plant genomics is lacking. [Method/Process] The input to deep learning applied to genomics is usually biological sequences and molecular phenotypes as predictor and target variables, respectively. We introduced the workflow from four views: input data pre-processing includes retrieval, coding, and splitting; model construction and training includes the selection of model architecture and hyperparameters; model evaluation and interpretability. Specifically, this paper introduces the background of deep learning approaches, including the latest graph neural networks; then it discusses two prominent issues in the intersection of genomics and deep learning with respect to gene characterization and protein characterization: 1) how to model the flow of information from plant genomic DNA sequences to molecular phenotypes; and 2) how deep learning models can be utilized to identify functional variation in natural populations? Specifically, the paper summarizes the current status of deep learning applications in related fields, which include deep learning and DNA and gene characterization research, interpretability of deep learning in genomics applications, graph neural networks in genomics, deep learning and genomic variation research, deep learning in protein prediction, ALPHAFOLD in protein prediction, deep learning and crop breeding research, and unsupervised learning in genomics and protein characterization. [Results/Conclusions] This article summarizes how traditional deep-learning algorithms, graph deep-learning, generative adversarial networks and interpretable AI are applied in current research in order to address these two problems. Finally, the prospects for deep learning in future plant genomics research and crop improvement are discussed. Overall, deep learning has provided better results than conventional methods in many genomics research directions, and the application of deep learning in genomics has yielded early applications of scientific and economic significance. Deep learning offers two distinct advantages: 1) end-to-end learning, with the ability to integrate multiple pre-processing steps into a single model; and 2) multimodal data processing capabilities that can handle extremely heterogeneous data in genomics. The advancement of deep learning has the potential to expand new research perspectives in genomics and crop breeding, and to facilitate larger-scale association studies in both phenotypic and genotypic genomics as algorithms become more accurate.

[Purpose/Significance] At present, most of the training models used in the research of news classification are non-active learning. There are common problems about these models, including data cannot be labeled immediately and the labeling cost is too high, which also hinders the analysis of agricultural news. Especially because of the explosive growth of news data in the network era, it is more difficult to label data, train supervised text classification models, and screen relevant news in the field of agriculture from diversified online news sources. In order to solve this problem, the most commonly used pool based active learning or deep active learning technique is used to select more valuable and representative data from unlabeled data for manual labeling, and construct labeled data sets to improve the efficiency and effect of news classification and agricultural news mining. [Method/Process] The commonly used machine learning models for text classification, such as random forest classifier, polynomial naive Bayes classifier and logistic regression classifier, were combined with the active learning method with the lowest confidence to analyze the effect, and the BERT model was combined with the three sampling strategies of discriminative active learning, deep Bayes active learning and lowest confidence for deep active learning training. On the news corpus of 19 847 samples crawled and cleaned by crawler technology from Sina and other news websites, aiming at screening agricultural related news from diversified news samples of various topics, the iterative experiment of adding 30 samples per round was tested to check the improvement effect of F₁ score under various method combinations with the increase of the number of annotation. In addition, the representativeness and diversity of the samples selected by the sampling function of each method in the deep active learning method of the BERT model were compared, so as to understand the characteristics of each strategy and provide inspiration for the selection and improvement of Al strategy in the future. In addition, this paper also analyzed how much labeling cost can be saved by using the proposed method. [Results/Conclusions] When comparing a variety of machine learning models, it is found that although the gradient boosting tree and support vector machine classifier have high accuracy, they are not suitable for active learning because of their low efficiency in text data processing of large-scale high-dimensional data. After combining other machine learning models and the BERT model and training text models with the corresponding active learning or deep active learning methods, it is found that the application of active learning method can significantly improve the training process of each model. Among them, the BERT model, combined with discriminative active learning sampling function, has the best news text classification effect and the lowest annotation data requirements. The representativeness and diversity of the samples selected by discriminative active learning sampling function are also the highest, which explains the source of the advantages of this method. It can also be found that for the same task model, the higher the accuracy of classification is required, and the active learning method can save more annotation cost than non-active learning.

[Purpose/Significance] With the development of data-intensive scientific research paradigm, the effective discovery and reuse of scientific data is of great significance to the sharing of research results. In order to promote scientific data utilization and improve the scientific value of data resources, the international academic community proposed the FAIR principles in 2016, aiming to promote data openness, which has so far attracted widespread attention from scholars at home and abroad. Many exploratory studies have been carried out around the FAIR principles, such as GO FAIR, and RDA, CODATA and other international organizations have been established to be devoted to open science and data sharing practices; the European Union, as the representative advocate of open science, has strongly promoted the implementation and promotion of the FAIR principles by establishing the European Open Science Cloud (EOSC) and introducing data policies, etc. The purpose of this paper is to provide a systematic and comprehensive overview of the academic achievements of FAIR principles, with the aim of providing references for the promotion and implementation of the principles and scientific data management practice in China. [Method/Process] This paper combines literature research and online research, and through content analysis, provides an in-depth analysis of literature on FAIR principles in terms of content interpretation, assessment methods, domain application, discipline application and regional application. It includes five perspectives of organization guarantee on FAIR principles, four basic principles of FAIR, practical exploration, disciplinary implementation and regional practice of FAIR principles to review the research progress and implementation of FAIR principles at home and abroad, focusing on exploring the practical value of FAIR principles, and systematically summarizing the research status and existing achievements. [Results/Conclusions] International research on FAIR principles focuses on theory, implementation strategies, and evaluation methods, For example, at the level of FAIR content, the four basic principles have been explored and discussed in practice; at the level of implementation strategies, many stakeholders have taken different measures to ensure the implementation of FAIR principles, and different organizations have developed evaluation methods. In addition, FAIR principles have been explored in the fields of medicine and other disciplines and are widely used in Europe and other regions. In contrast, domestic research is still in its infancy and national policy support and follow-up by relevant organizations is needed for further study and implementation of FAIR principles in China.

[Purpose/Significance] At present, the academic evaluation system has begun to take shape, but the existing evaluation methods still have their own disadvantages, which bring some disputes to the scientific nature of academic evaluation. Scholars at home and abroad analyzed the current academic evaluation activities. Many theoretical assumptions were put forward to promote the development of academic evaluation. At the same time, in order to make academic evaluation more scientific, scholars put forward improvement plans from the technical aspect. However, only a specific solution to a problem in academic evaluation is proposed, and the research on academic evaluation system based on blockchain is still blank. Therefore, the paper asks this question: Is it feasible to apply blockchain technology to academic evaluation? How to build an academic evaluation system based on blockchain technology? [Method/Process] Regarding the traceability characteristics and timestamp technology of blockchain data, the peer-to-peer transmission and distributed consensus algorithm of this technology were used to circulate academic coins in the academic evaluation system, and the academic currency equivalent to academic value platform was established. Using the imtamable characteristics of blockchain data and the timestamp technology, the electronic tag of academic research is established, so as to achieve comprehensive monitoring of all steps of academic research and ensure the objectivity of academic research at the data source. Combined with block chain technology, based on block chain academic currency circulation system, and with academic currency equivalent value platform set up, block chain was applied to the academic evaluation, which puts literature related information in the block in the chain store, and at the same time encourages users to do literature review, and implement comments thumb up for academic currency of incentive mechanism, to improve the impartiality of the academic evaluation. [Results/Conclusions] The system is capable of effectively improving the problems of scattered downloads and usage and incomplete statistics in the evaluation of influence, and enhancing the fairness of peer review after publication; At the same time, it can effectively realize copyright protection and scientific research reproduction. The academic coin circulation system based on block chain can be applied to scientific evaluation, copyright protection, user credit and other scenarios, which can provide beneficial inspiration for the future application of block chain in academic evaluation. For example, the circulation rules and incentive mechanism of the academic currency system proposed in this paper are not comprehensive enough. It is necessary to further improve the circulation rules and research ideas of the academic currency system.

[Purpose/Significance] A library portal website is related to the function and social image of a library. Library website construction is also one of the tasks of the development of library cause in China during the 14th Five-Year Plan period. However, the website construction level of small and medium-sized libraries at and below the county level in China is relatively low, and effective measures are urgently needed to improve it. To this end, Japanese libraries' experience is worth learning about. Therefore, we take the websites of the municipality libraries in the Tokyo Metropolitan Area as the research objects to explore the information construction and design concept of the websites of small and medium-sized libraries in Japan, so as to provide suggestions for the improvement of the website of grass-roots libraries at and below the county level in China. [Method/Process] Firstly, we analyzed and expounded on the construction principles of small and medium-sized library websites in Japan. Then we conducted a network survey on the website construction of the libraries in Tokyo and summarized the information construction and design concept of the websites from the three aspects of website structure, UI, and function design. In terms of website structure, we investigated the design of navigation, retrieval (simple retrieval, advanced retrieval, and bibliography recommendation), regional data area, juveniles' pages (children's page, teenagers' page), mobile terminal, etc. In terms of UI, we investigated interaction design, color matching, visual identity, etc. In terms of functions, we focused on exploring the accessibility services, business support services, and education support services of the websites based on the concept of a "problem-solving library" in Japan. [Results/Conclusions] The structural design of Japanese small and medium-sized libraries' websites is oriented to the convenience of the people and pays attention to the needs of children and the regional cultural characteristics. The UI design is simple and intuitive, the color matching is soft, and the visual identity is clear. Their functional design focuses on problem-solving, especially in accessibility, business, and education support services. Therefore, according to the theory of responsible innovation, this paper puts forward four suggestions for the website construction of small and medium-sized libraries in China: (1) to be focused on local characteristics and strengthening the construction of local information resources; (2) to take users as the center to meet the diversified cultural needs; (3) to optimize basic services and improve service convenience and ease of use; (4) to rely on the general-branch library system to ensure operation to achieve sustainability.

[Purpose/Significance] The modern courier industry is developing rapidly and the demand for courier services is increasing. A large amount of personal information is involved in parcel waybills, and even if privacy waybills are used, there is still a risk of personal information exposure. This study attempts to examine whether invisible face slips in the courier industry can effectively protect users' personal information in accordance with the Personal Information Protection Law implemented in November 2021 in China, and to propose feasible suggestions for the protection of personal information within the courier industry. [Method/Process] This study selected 9 major courier companies in China as the research object, and investigated and compared the courier agreement terms and privacy waybills by collecting each courier company. By comparing the terms of express agreement of each courier company, we investigated the perception and definition of personal sensitive information of the 9 major courier companies, and studied the general understanding and difference between personal information and personal sensitive information within the courier industry. By further comparing the invisible face sheet standards of each courier company and within the courier companies, we investigated the differences in the anonymization of parcel waybills and studied the implementation of invisible face sheet standards and differences in the courier industry. After completing the above investigation, the effectiveness of the courier industry in terms of personal information protection was compared by designing a personal information protection index and analyzing the variability among courier companies and within courier companies. [Results/Conclusions] The study found that there is no unified standard in the implementation of privacy waybills in China's courier industry, and the perception of sensitive personal information has not been unified, resulting in the phenomenon of different standards of privacy waybills among courier companies. The phenomenon of different invisible face sheet standards exists not only among courier companies, but also within courier companies. The specific problems are revealed by the different regionalization of invisible face sheet standards within courier companies, and this study classifies the degree of personal information protection into 3 levels, namely excellent, good and dangerous, with the excellent region of East China. In response to the phenomenon of different invisible face sheet standards in the courier industry, China needs to develop a special courier industry standard to reduce the problem of exposing users' personal information and form an information association with the different invisible face sheet standards implemented in parcel waybills. The study also found that the degree of impact and sensitivity of personal information varies in the use scenarios of different industries, and the Personal Information Protection Law has a broad definition of personal sensitive information, which requires the development of special definition standards depending on the circumstances of each industry to further strengthen the protection of personal information in China. This paper puts forward 3 feasible suggestions to address the above issues: first, to clarify the personal sensitive information involved in privacy waybills; second, to build a unified privacy waybill standard within the express industry; and third, to strengthen the implementation of promoting the unified standard.

[Purpose/Significance] Rare ancient books are an important carrier of Chinese culture. At present, the focus of ancient books research has been on the original and regenerative protection, while the conservation of inheritance is relatively lagging behind, and the role of the promotion of ancient books reading has not been fully played. The emergence of virtual reality (VR) technology provides a new idea and method for the intelligent transformation of ancient book reading promotion, which is of great value to the construction of a library's intelligent knowledge service system and the innovation of the service mode. With the help of VR technology, the intelligent reading promotion of ancient books shows the characteristics of "narrative presentation scene", "reading service immersion" and "reading space humanization". [Method/Process] In order to provide reference for the research and development of intelligent reading in the new era, this paper first discusses the concept "gene" of intelligent reading of ancient books enabled by the VR technology, and summarizes the application scenarios of the VR technology in the promotion of intelligent reading of ancient books, combined with the National Library of China's specific case on "VR panorama cultural classics". Then the application path of VR technology was put forward enabling the intelligent reading and promotion of ancient books. To this end, this paper first summarizes the concept "genes" of intelligent reading promotion from a macro perspective, and then analyzes the application scenarios of VR technology in the promotion of intelligent reading of ancient books based on the investigation of the application cases of "VR + ancient book reading promotion". It analyzes the construction practice of "VR Panorama Cultural Classics" in the National Library of China from four perspectives of content narrative, visual presentation, empathy mechanism and communication strategy, to construct a "VR and ancient books" reading promotion path. The content narrative of "VR panoramic cultural books" should focus on activating characteristic collection's historical materials and recording key historical moments. In the aspect of visual presentation, it should pay attention to excavating the elements of characteristic cultural scene and reconstructing the landscape scientifically and rigorously. In terms of the mechanism of empathy, the focus should be put on the use of the block type structure to present the history panorama, and the use of typical characters to compose stories. When it comes to communication strategy, it should pay attention to jointly creating a new pattern of "library +" by combining online and offline multi-screen media. [Results/Conclusions] A library should first pay attention to narrative expression, and the control of the narrative logic of ancient book stories. Second, the library should pay attention to scene perception and explore the presentation mode of panoramic vision threshold. Third, attention should be paid to top-level design and giving play to the guiding force of ancient books' cultural engineering. Fourth, the library should pay attention to cross-border cooperation and adhere to the concept of library reader service. In the promotion of reading ancient books, full consideration should be given to the particularity of ancient books resources and existing reading difficulties. In the process of content dissemination, the focus should be on contructing a model through ring upon ring decomposing and progressively thinning method, and taking targeted measures from the aspects of top-level design, strengthening demonstration, building benchmarking, series construction, cross-border cooperation, building platforms, strengthening perception and strengthening ethical supervision, so as to promote the intelligent reading and promotion of ancient books.

[Purpose/Significance] Based on the needs of professional degree education and innovation in China and the relevant policy background of industry-university-research collaborative education, this paper investigated and analyzed iSchools' library and information science (LIS) industry-university-research projects in the United States. The content analysis helps to optimize the training system of LIS professionals in China, and provides theoretical reference for promoting the active participation of industry-university-research in social and economic development in China. [Method/Process] Based on the network survey method, this paper selected 35 representative LIS industry-university-research projects in 20 iSchool colleges and universities in the United States, analyzed in detail from the aspects of participation subject, funding source, research field, research object, social demand and evaluation system, summarized its content and characteristics, and put forward suggestions for the development of the collaborative education of LIS industry-university-research in China. [Results/Conclusions] In order to promote the in-depth development of industry-university-research education, when conducting industry-university-research projects, LIS institutions in China should support multi-agent participation in industry-university-research project collaboration, innovate cooperation forms, and encourage multi-agent participation in 'joint training' of professional practice. In the direction of project research, the training unit should pay more attention to the interdisciplinary nature and integration, broaden the research field of the project, so as to enrich the form of innovation cooperation. China's LIS training units should also broaden the funding channels, strengthen project publicity, select high-quality cooperative researchers to jointly declare or undertake scientific research projects, and use the expected benefits of results to extend the economic benefit chain of cooperative units. In the project research, the cooperation subjects should be oriented by the current social needs, optimize the training system of professional education, and strengthen the training of students-oriented professional practical skills by using the advantages of regional policies, geography and cultural heritage of colleges and universities and partners. The educational subjects mainly based on the Teaching and Instruction Committee of Library and Information Science in China should actively play the advantages of multi-subjects of industry-university-research to enrich the level of industry-university evaluation, encourage multi-form evaluation and awards, and ensure the application, social benefits and innovation of project results, so as to promote the long-term development of industry-university-research in LIS in China.