[Purpose/Significance] Deep learning for language modeling is one of the major methods and advanced technologies to enhance language intelligence of machines at present, which has become an indispensable important technical means for automatic processing and analysis of data resources, and intelligent mining of information and knowledge. However, there are still some difficulties in using deep learning for language modeling for technology development and application service in the library and information science (LIS) field. Therefore, this study systematically reviews and reveals the research progress, technical principles, and development methods of deep learning for language modeling, with the aim at providing reliable theoretical basis and feasible methodological paths for the deep understanding and application of deep learning for language modeling for librarians and fellow practitioners. [Method/Process] The data used in this study were collected from the WOS core database, CNKI literature database, arXiv preprint repository, GitHub open-source software hosting platform and the open resources on the Internet. Based on these data, this paper first systematically investigates the background, basic feature representation algorithms, and representative application development tools of deep learning for language modeling, reveals their dynamic evolution and technical principles, and analyzes the advantages and disadvantages and applicability of each algorithm model and development tool. Second, an in-depth analysis of the possible challenging problems faced by the development and application of deep learning for language modeling was performed, and two strategic approaches to expand their application capabilities were put forward. [Results/Conclusions] The important challenges faced by the application and development of deep learning for language modeling include numerous parameters and difficulties to adjust accuracy, relying on a large amount of accurate training data, difficulties in making changes, and the intellectual property and information security issues. In the future, we will start from two aspects of specific domains and feature engineering to expand and improve the application capabilities of deep learning for language modeling. Specifically, we focus on consideration of the collection and preparation of domain data, selection of model architecture, participation of domain experts, and optimization for specific tasks, in order to ensure that the data source of the model is more reliable and secure, and the application effect is more accurate and practical. Moreover, the strategic methods for feature engineering to expand the application capabilities of deep learning for language modeling include selecting appropriate features, feature pre-processing, feature selection, and feature dimensionality reduction. These strategies can help improve the performance and efficiency of deep learning for language models, making them more suitable for specific tasks or domains. To sum up, LIS institutions should leverage the deep learning for language modeling related technologies, guided by the needs of scientific research and social development, and based on advantages of existing literature data resources and knowledge services; they should carry out innovative professional or vertical domain intelligent knowledge management and application service, and develop technology and systems with independent intellectual property rights, which is their long-term sustainable development path.

[Purpose/Significance] ChatGPT, an artificial intelligence chat robot based on transformer neural network that was developed by OpenAI, has now leveled up to GPT-4 and GPT-4 Turbo. solves the problem of peer-to-peer response in human-computer interaction with its powerful natural language processing ability. Its abilities of information retrieval, language organization, logical reasoning and text induction meet the needs of library users, which makes the transition from a traditional library to a smart library with a return to knowledge service possible. The paper focuses on the development trends in the library field in the era of artificial general intelligence. [Method/Process] Based on the background of the library development in the computer age, the paper reveals the dilemma of library knowledge service through retrospective and comparative methods, pointing out the endogenous growth value of ChatGPT's high-quality analysis, demonstration and generation ability with multi-modal and full-scene deep interaction for smart library services. Library service presents a new mirror image from the perspective of ChatGPT: intelligent retrieval covers all traditional service, while the focus of information consultation, subject service and reading promotion shifts and the service margin becomes blurred. Information consultation retains the necessary share of traditional consultation, and becomes professional and focused; subject service moves forward to the frontier of all-disciplinary field; reading promotion realizes the accurate docking between the collection and the target population. The potential risks and challenges posed by ChatGPT include copyright risk and secondary institutional copyright risk with plagiarism, simulated viewpoint and the exposure of data. This urges the society to regulate the development behavior of technology companies through legislation, and requires third-party supervision and local area networks to ensure security. [Results/Conclusions] On the basis of saving human resources and management costs, ChatGPT accelerates the optimization of traditional library services, improving service quality and efficiency. It may facilitate the transfer and reorganization of traditional library services, but this does not necessarily lead to the end of the library industry. On the contrary, it makes possible for the library industry to return to the knowledge service with the blessing of new technologies. It is inevitable to consolidate divisions, streamline personnel and rebuild teams, but the process may not be perceptible or even dramatic. The service standards and evaluation system of the library industry change from borrowing and returning books, click rate and pageview to the depth and efficiency of teaching and scientific research participation. Knowledge service becomes clear and conscious. The threat of digital books and electronic reading has been greatly reduced, making it easier for the library industry to confirm its position in the social and historical development.

[Purpose/Significance] At present, studies on online health information search at home and abroad mainly focus on the analysis of influencing factors of online health information search, the analysis of characteristics of online health information search, and the analysis of the action path between online health information search and health anxiety, etc. Few scholars focus their research objects and perspectives on offline medical treatment behavior. Moreover, there is insufficient research on the effects and mechanism of online health information search on offline medical treatment behavior. This paper aims to analyze the characteristics of public online health information search and explore the factors that affect public online health information search on offline medical care seeking behavior. [Method/Process] Based on the S-O-R theory, this paper proposed relevant hypotheses and constructed a correlation model between online health information search and offline medical care seeking behavior. Taking college students as the main research object, based on the maturity scale of domestic and foreign authorities, four variables of online health information search, risk perception, physical alertness and offline medical treatment behavior and 13 operable observation variables were designed. SPSS25.0 software was used to conduct reliability and validity analysis, correlation analysis and factor analysis of 204 valid questionnaires collected. Amos25.0 software was used to construct a structural equation model to verify the hypothesis and the Bootstrap method was used to test the mediation effect. [Results/Conclusions] The research findings: 1) public online health information search significantly affects offline medical seeking behavior; 2) public online health information search significantly affects risk perception; 3) public physical alertness significantly affects offline medical care seeking behavior; 4) risk perception and body alertness mediate between public online health information search and offline medical care seeking behavior. Based on the study conclusions and the results of interviews with some surveyors, two suggestions have been put forward. On the one hand, it is necessary to collect online resources, drive offline traffic, and help the growth of the offline medical market. On the other hand, it is necessary to build a health index, establish a nationally unified and credible evaluation index system for the health of urban residents, index evaluation of residents' health status, and predict disease manifestations. Most of the subjects in this study are college students, and the questionnaire distribution scope is limited, so the research results are not representative enough. Additionally, the scope of the research object will be further expanded to include the middle-aged and senior subjects to enhance the persuasiveness of the research conclusions.

[Purpose/Significance] This study aims to complement the shortcomings existing in the current research, such as the limited research on the micro-level of enabling rural revitalization and the unclear impact of knowledge services on rural revitalization, in order to strengthen the understanding of the mechanism and path of enabling rural revitalization by agricultural intelligent knowledge services, and to provide theoretical reference and practical solutions for better enabling rural revitalization by the knowledge services and better developing agricultural intelligent knowledge services. [Method/Process] This paper first reviews the literature to understand the current research status of enabling rural revitalization and knowledge services affecting rural revitalization, and then determines the research focus. Based on the three perspectives of data, application and service, this paper explores the enabling basis of agricultural intelligent knowledge services, analyzes the enabling mechanism from the three aspects of agricultural production, farmer life and rural governance, clarifies the necessity and possibility of agricultural intelligent knowledge services enabling rural revitalization, and constructs the internal mechanism of the knowledge services enabling rural revitalization. Moreover, the predicament of agricultural intelligent knowledge services enabling rural revitalization is analyzed from the internal and external aspects, and the performance and impact of these dilemmas are analyzed. Finally, according to the analysis results, the optimization path is proposed from the three dimensions of enabling demand, scene and effect to alleviate the problem. [Results/Conclusions] Agricultural knowledge services have entered the intelligent stage, presenting the characteristics of intelligence, personalization and precision, and have established a good foundation for enabling rural revitalization in terms of data, application and service. By enabling the development of new production modes, the transformation of farmers' multiple identities, and the equal participation of multiple subjects in governance, digital intelligence momentum will be injected into the comprehensive promotion of rural revitalization. However, from the internal perspective of agricultural intelligent knowledge services, there are some shortcomings such as obvious user group bias and the lack of service content. From the external point of view of the knowledge services, there are the challenges of unbalanced infrastructure and low user literacy. These deficiencies and challenges limit the enabling role of agricultural intelligent knowledge services in rural revitalization, which will limit the maximum play of the enabling effects. Accordingly, this paper provides the corresponding solution. First, in-depth research could be carried out on rural revitalization stakeholders and strategic content to clarify the empowerment needs. Second, integrated service platform could be built to integrate and empower scenarios. Third, the promotion model could be constructed to enhance the enabling effect to alleviate the empowerment dilemma.

[Purpose/Significance] Mobile reading has become an important way to acquire knowledge. As an important institution providing public reading services, public libraries have launched various forms of mobile reading services utilizing new technologies to meet changes of user needs. Compared with market-oriented mobile reading services, a public library's mobile reading services have certain disadvantages, such as poor user experience, limited functionality, lack of reading aids, and weak social functions. By studying the market-oriented operation of mobile reading services, it is found that the common parts of mobile reading services can provide a certain reference for the construction of a library's mobile reading services and improve the user experience of the library's mobile reading services. [Method/Process] This paper mainly adopts a survey method. It selects representative mobile reading apps in the market, including iReader, Migu Reading, Shuqi Novel, WeChat Reader, and NetEase Snailreader. The mobile reading function is divided into five parts: bookshelf, bookstore, book details page, personal center, and reader, and the functions of each part are fully investigated. Due to the fact that reading is the core function of mobile reading services, we focused on investigating the reader functions of various apps and divided all the functional points of readers into six categories, namely reading content positioning, reading social and interactive, reading assistance, personalized settings, book management, and other functions. A summary of the basic functional points of mobile reading apps surveyed is provided, which serves as a foundation for the subsequent proposal of a functional framework for mobile reading services. [Results/Conclusions] The focus is on proposing the front-end and back-end functional framework of the mobile reading services. Based on the investigation of five representative mobile reading apps and the author's years of experience in mobile reading services, the basic functional framework of the mobile reading service front-end is proposed. Based on the current mainstream information technology and user reading needs, an extended functional framework has been proposed, including five aspects: enhancing user experience design, enhancing content recommendation function, creating multi-dimensional knowledge associations, strengthening the construction of reading aids, and designing smart reading services. This provides a reference for the development of mobile reading products by libraries. Moreover, the article proposes a back-end functional framework for mobile reading services from the perspectives of user behavior management, resource publishing and management, statistics and big data analysis, copyright management, and interface services, which fully considers the needs of different users and the publishing standards of mobile reading service resources. In addition, the basic functional framework of mobile reading service front-end is mainly proposed based on the research of existing mobile reading products. Although some expansion suggestions have been made, there is still a lack of practicality in smart libraries. There are relatively few reference materials for the back-end functional framework, mainly based on the author's years of work experience and a small amount of network resources. Further research is needed in terms of applicability.

[Purpose/Significance] In the era of big data, the volume of data is growing at an exponential rate. One of the most prominent areas affected by this growth is the field of agriculture. The use of agricultural knowledge graphs, which serve as key infrastructures for managing agricultural knowledge, has expanded significantly. However, as the number of nodes and relationships within these graphs increase, so too does their complexity. This complexity gives rise to new challenges in training and representing such large-scale knowledge graphs. It is therefore of great significance to investigate methods for speeding up the embedding process of agricultural knowledge graphs, while preserving their structural integrity and minimizing resource consumption. This research embarks on a novel exploration to address this issue. It stands out from previous studies by concentrating on a hierarchical representation model for agricultural knowledge graphs. The potential impacts of this research on propelling the advancement of the field and on addressing significant real-world problems are substantial. [Method/Process] To confront this challenge, we propose a hierarchical representation model for agricultural knowledge graphs rooted in the HARP framework. Our model leverages the inherent hierarchical features of the agricultural knowledge graph. It incorporates an improved random walk strategy based on relational paths to semantically model relationship objects within the agricultural knowledge graph. This innovative approach effectively retains the hierarchy and asymmetrical relationship structure of the nodes in the graph, setting our work apart from previous research. The validity of our proposed model is fortified by a strong foundation of theoretical and empirical evidence. [Results/Conclusions] Our experimental results reveal several key findings. First, the hierarchical random walk with path (HRWP) model using the LEIDEN algorithm can preserve the spatial structure more effectively and converge more quickly to the maximum modularity, in comparison to the HARP framework. Second, the fusion model employing HRWP takes less training time than the total training time of both models combined, without significantly affecting the time complexity of the original algorithm. Third, we observed that when traditional algorithms are integrated with HRWP, there is an average improvement of 2% across various indicators, with a substantial enhancement in non-neural network models. Therefore, our proposed model not only accurately represents the agricultural knowledge graph but also effectively reduces the training time. Despite the promising outcomes of our study, there remain areas of potential improvement. One such area is the need for a more detailed discussion on the hierarchical nature of relationship objects in future research. This provides potential avenues for future exploration in this field. The findings of this research carry profound implications for the development of agricultural knowledge management systems, offering an effective approach to handle the burgeoning complexity of knowledge graphs.

[Purpose/Significance] The development of open-source software and its system in China is relatively slow compared with the United States. However, given the current development status and needs, it is urgent to establish a sound institutional mechanism to provide feasible solutions for the development and protection of open-source software, and leave sufficient space for addressing potential risks. [Method/Process] Comparative research methods were used to examine extraterritorial systems and reflect on the current situation in China. By examining the reasons for the emergence, and the controversies formulated of the copyright licensing investigation system in the United States, and analyzing the content of the system, it can be seen that the system aims to investigate potential infringement of open-source software, and strives to balance the development and protection of open-source software. On the contrary, there is a gap in China's legislation, and the legal status of open-source software copyright has not yet been established, let alone specific institutional provisions on the implementation subject and authority design. [Results/Conclusions] It is necessary to learn from the copyright licensing investigation system in the United States. Based on the actual situation in our country, we can establish the legal status of copyright protection for open-source software based on the lawsuit of intellectual property infringement in the Civil Code of the People's Republic of China, referring to the open-source software protection model established in the Welte v. D-Link case and Harald Welter v. SiteCom case in Germany, by clarifying the legal effect of open-source software, incorporating copyright protection, allowing infringement lawsuits, and stipulating the right to request information from the rights holder. In terms of specific institutional provisions, the subject of copyright licensing investigation rights can be temporarily determined as the copyright administrative department at or above the prefecture (city) level. After the development of relevant public welfare organizations such as the Open Source Foundation is mature, some or all of the powers may be delegated to third-party public welfare organizations according to the situation. Furthermore, the rationality and legality of the investigation and the application of qualified sponsors are taken as the premise of the investigation, the type of equipment under investigation is limited to the legally obtained equipment, the investigation of potential infringement of open source software is regarded as the only investigation object, and the proper handling and protection after the investigation is taken as the requirement to ensure the legitimacy of the investigation, in order to complete the authority design of the investigation subject, and the localization construction of the copyright licensing investigation system will be achieved in the end.

[Purpose/Significance] In order to explore the technological gaps in Chinese im-portant agricultural fields and predict the future trends of these gaps, this study investigates technology opportunity discovery in the embryonic and developmental stages from the per-spectives of technology gap discovery and technology fusion opportunity discovery, provid-ing consultation and suggestions for decision-makers on the technology development op-portunities for technology innovation. [Method/Process] First, TextRank method was used to mine information in abstracts of papers and patents in this paper, which is a key sentence embedding method. The sentence vector clustering method was applied to extract topic sen-tences of papers and patents. Second, comparative analysis of topic clustering was used to detect technology gaps. Third, semantic similarity networks and classification similarity networks were used to discover the theme directions, which are likely to develop into cross-domain research areas with these technology gaps. [Results/Conclusions] The experi-mental results indicate that the proposed method can identify technological gaps. Combined with expert analysis, the experimental results can show the current development status and predict the trends of genomics technology in the field of animal husbandry and veterinary medicine. At the same time, this study can provide methodological and data support for genomics technology think tanks in the field of animal husbandry and veterinary medicine in China. Specifically, China has published a large number of papers and patents, but the tech-nical architecture layout is not as complete as foreign countries. The topics of Chinese papers are more complete than those of Chinese patents. In addition, China lacks sufficient basic research support in the integration and association of multi-omics, and the technical conditions are also incomplete. The field of genetically modified (GM) breeding technology is also recognized as a technological gap in China. In addition, it is possible that GM breeding and whole genome association analysis, multi-omics integration and viral genome analysis of livestock and poultry will become new technological fusion points in the future. There are still drawbacks in this study: It still takes time and manpower to manually analyze and interpret the relationship between scientific papers and technological patents. In the future research, more automated methods will be designed to construct correlation comparison methods between two data objects. Additionally, there is still room for improvement in expert interpretation of clustering themes. In the future, more data can be considered to add label information, reducing manual annotation work while providing the possibility of increasing quantitative accuracy in the result validation section.

[Purpose/Significance] The collaboration between the instructor of a general-purpose course "Writing and Communication" and information literacy (IL) librarian teachers at Southern University of Science and Technology provides a new embedded teaching cooperation scheme for an undergraduate writing course. This new model can help college academic librarians design more active ways to teach embedded IL class in general-purpose writing courses for freshmen and sophomores against the background of cultivating "top innovative professionals". [Method/Process] Based on the teaching collaboration between the librarians and the instructors in the past five years, we explored and analyzed effective and integrated teaching models of embedded IL education. Focusing on assessment and student's learning outcome, we constructed an embedded IL teaching evaluation model to comprehensively evaluate student's learning outcome, which includes both direct and indirect evaluation. The direct assessment includes five evaluation indicators: information awareness, information inquiry and access, information evaluation, information management and utilization, and information regulation and security. The indirect evaluation is a "librarian-writing instructor" dual assessment based on students' course papers, in which the librarian and the writing instructor rate students' IL skills and writing skills independently and then compare the assessment results. The librarian scrutinizes students' work under the rubric of search strategy, resource selection and reference standardization; and the writing instructor examines the critical thinking, essay structure and writing style of students' papers. The comparison of the scores on writing and information use of students' papers provide quantitative evaluation on the correlation of students' writing and information skills. [Results/Conclusions] There is a positive correlation between IL and writing performance. College freshmen differ from senior students in IL competence in many ways due to their lack of information training and of limited exposure to various academic scenarios. Thus, besides teaching basic searching skills, instructors should also help students build up information awareness and competence by situating IL teaching in specific writing tasks in an integrated teaching model. By collaborating with writing instructors, the embedded information session is more contextualized, and therefore a "process-based" IL development can be demonstrated to students. Other new pedagogical approaches such as flipped classroom also help to enhance teaching and learning.