[Purpose/Significance] Traditional Chinese medicine (TCM)texts contain a wealth of domain-specific knowledge that can provide guidance for accurate diagnosis and effective disease prevention and treatment.This paper provides a systematic review of the research on named entity recognition (NER)in TCM. [Method/Process] From the characteristics of TCM texts, the paper discussed the challenges of TCM texts NER including knowledge system, corpus construction, and technical algorithms aspects.It sorted out the terminological standards, entity types, and annotation principles and methods available in the construction of TCM texts NER corpora.The paper also reviewed the general framework, common methods, and recent trends of TCM texts NER techniques and summarized the evaluation indicators. [Result/Conclusion] Future research can be conducted in the following directions: establishing annotation standards and constructing high- quality datasets at the corpus level, exploring data optimization for small sample problems, developing recognition models for complex entities, and enhancing model interpretability at the algorithm level to improve the effectiveness of TCM NER.

[Purpose/Significance] The automatic segmentation of traditional Chinese medicine (TCM)epidemic ancient texts and the mining of drug law based on segmentation results in the context of digital humanities are important ways to promote the activation of ancient book knowledge and clinical diagnosis and treatment support. [Method/Process] The research proposed a framework for mining drug laws based on automatic segmentation in TCM epidemic ancient texts.The framework included a data collection layer, a sequence annotation layer, an automatic segmentation layer, and an application service layer.Through four layers of collaborative linkage, the automatic segmentation of epidemic ancient texts and the mining of drug laws were ultimately achieved. [Result/Conclusion] The empirical results show that the F of the BiLSTM-CRF based automatic segmentation of TCM epidemic ancient texts proposed in this article reaches 92%.Based on the segmentation results, the mining results of various dosage forms, commonly used TCM, and commonly used drugs in the formula provide support for epidemic prevention and control, diagnosis and treatment decision-making assistance in the future.

[Purpose/Significance] This research realizes the dynamic organization and visualization of ancient literatures of traditional Chinese medicine (TCM)epidemics by constructing dynamic knowledge graph, and mines the hidden knowledge of ancient literatures of TCM epidemics. [Method/Process] The study proposed a knowledge element semantic description model of ancient literatures of TCM epidemics with dynamic elements, and based on this model and deep learning techniques the research took the warm disease school as an example to perform knowledge extraction and construct dynamic knowledge graph, and combined the knowledge computing method to analyze the evolution of the prescription and medication laws of the ancient books of the warm disease school as well as to analyze the laws of discernment and development. [Result/Conclusion] The results show that with the time, the warm disease school gradually prefer using sweet, bitter and cold herbs, the description of epidemic disease symptoms are relatively consistent in each period, and the theories of syndrome differentiation are increasingly expanded.The study reveals the evolutionary development laws of TCM epidemic knowledge and provides new ideas for the discovery of TCM knowledge and other related researches.

[Purpose/Significance] From the perspective of complex systems, combining the synergistic theory and role identity theory to reveal the internal mechanism of knowledge mobilization in community of practice, it is of great theoretical value and practical significance to solve the problems of role positioning and identity cognition in order to realize the efficient operation of knowledge mobilization. [Method/Process] According to relevant research results, this paper summarized five knowledge mobilization roles, proposed five complex driving forces, explored the role of identity synergy and role recognition in promoting knowledge mobilization, and constructed a dynamic model of knowledge mobilization and an operational mechanism. [Result/Conclusion] There are four"core roles"and one"marginal roles"in knowledge mobilization activities.Strengthening role cognition can promote organizational role identification.Identity synergy helps to enhance the adaptability of complex environment; The knowledge mobilization from the perspective of complex system realizes the value emergence under the flexible combination of five complex dynamics; A multitude of"micro-mobilization"phenomenon, from small and large, from local to overall, are key to promote the sustainability of knowledge mobilization.

[Purpose/Significance] Water disasters happen frequently all over the world in recent years, which lead to huge casualties and property losses in China.The reason is the delayed information transmission, information distortion and insufficient utilization in all stages of water disaster emergency response.Therefore, it is necessary to explore water disaster emergency information from multiple perspectives, especially the whole process of collecting, processing and generating all kinds of information, so as to serve the efficient response to water disaster events. [Method/Process] Aiming at the characteristics of water disasters and decision-making requirements at various stages, based on the similarity between biological fermentation and information generation, the theory of Knowledge Convening and Fermenting Model was innovatively introduced to reconstruct the fermentation factor mechanism and fermentation function mechanism of emergency information fermentation in water disaster emergency management, and then a water disaster emergency information generation model was constructed, and the applicability of the model was verified through empirical analysis. [Result/Conclusion] From the perspective of Knowledge Convening and Fermenting Model, this paper puts forward the improvement strategy of emergency information management from three aspects according to the water disaster emergency information generation model.The water disaster emergency information generation model constructed can significantly improve the efficiency and effect of water disaster emergency information management.

[Purpose/Significance] In the context of the big data era, the complex information environment has generated great challenges to intelligence work, and the establishment of a rapid and direct intelligence delivery system is an inevitable requirement for intelligence work. [Method/Process] This paper proposed a new intelligence work integration model sprouting from the challenge of big data—the intelligence work model guided by the idea of intelligence middleware.By analyzing the similarities and differences among intelligence middleware computer middleware, cloud computing middleware and Internet of Things middleware, etc., the conceptual connotation and function of intelligence middleware were clarified, and the intelligence service process and empirical research under the idea of intelligence middleware were exemplified, which preliminarily confirmed the feasibility of the idea. [Result/Conclusion] The study shows that the intelligence middleware can improve the efficiency of intelligence work through the cohesion of dispersed academic research results.

[Purpose/Significance] This paper aims to explore the evolution path of China's AI industrial policy in the uncertain environment,clarify the change and future trend of the focus of China's AI development process,and provide a reference for the further improvement and high-quality development of AI industrial policy. [Methodology/Process] The article combined the methods of bibliometrics,network analysis,co-occurrence analysis and thematic structure evolution to analyze the number of AI industrial policies,co-publishing institutions and thematic structure evolution from three perspectives respectively. [Results/Conclusions] The study finds that the number of AI industrial policies issued in China has increased year by year;the central and local governments are coordinated and unified;and the policy themes are increasingly rich,but there are still shortcomings such as low innovation of AI technology,lack of policy balance,and lack of international cooperation.Future AI industrial policies need to be further improved and optimized in terms of technological innovation,education system,employment issues and international cooperation.

[Purpose/Significance] The quantitative analysis of policy diffusion is the application of quantitative analysis of policy literature in the field of policy diffusion, representing an interdisciplinary research direction.This paper systematically reviews the latest advancements in quantitative policy diffusion research from three dimensions: theory, methodology, and application, which can guide the practice of analyzing the diffusion of dense policy texts. [Method/Process] The paper explained the conceptual essence, data characteristics, research framework, and application scope of policy diffusion quantification.At the methodological level, it is divided into two types: policy diffusion econometric analysis and policy diffusion text mining methods.At the application level, the main focuses were studies on the impact of policy diffusion, diffusion patterns, and diffusion structure assessments. [Result/Conclusion] Quantitative research on policy diffusion is developing rapidly, mainly focusing on the transfer and application in interdisciplinary fields.In the future, efforts need to be made in the development of specialized research tools, the innovation of integrative methods, and the expansion of data types.

[Purpose/Significance] Data security is the premise and foundation to ensure the development of new quality productive force.It is of great value to study the theoretical model, mechanism and practice path of data security to ensure the development of new productivity. [Method/Process] Firstly, based on the four dimensions of data security"subject-object-mediator-ring", a four-dimensional theoretical model of data security to ensure the development of new quality productive force was constructed by using the theoretical modeling method.Secondly, the system theory analysis method was used to analyze the input mechanism, transformation mechanism, output mechanism and adjustment mechanism of data security guarantee new quality productive force development.Finally, adhere to the combination of theory and practice, and it built a practical approach to data security to ensure the development of new quality productive force from four aspects: kinetic energy, foundation, talent and ecology. [Result/Conclusion] Data security generally covers the three major structures of workers, means of production and labor objects in the development of new quality productive force from the four dimensions of subject dimension, object dimension, carrier dimension and ring dimension.The guarantee system of data security for the development of new quality productive force includes four mechanisms: input mechanism, transformation mechanism, output mechanism and adjustment mechanism.It is urgent to construct the kinetic energy, foundation, talents and ecology of data security to ensure the development of new quality productive force from four aspects: maintaining scientific and technological data security, industrial data security, cultivating data security talents and strengthening data security governance.

[Purpose/Significance] Digital ecology can actively guide and promote the security of public data, provide a new vision for the solution of security problems in public data opening and utilizing, and help the sustainable development of public data opening and utilizing. [Method/Process] Based on the analysis of digital ecology theoretical framework, the public data security guarantee system was constructed. [Result/Conclusion] The public data security guarantee system consists of the public data security support system, the public data security infrastructure system, the public data security capability system, the public data security policy system and the public data security operation system.Then the relationship between public data opening and utilizing and security guarantee can be effectively balanced.

[Purpose/Significance] This paper aims to analyze the development trend of interdisciplinary by combining the external discipline attributes and internal topic evolution.It is helpful to capture the development frontier of the field and promote the development of interdisciplinary innovation. [Method/Process] Based on the overall research and breakthrough research of a target field, this paper firstly constructed the interdisciplinary co-occurrence network to analyze the evolution of external discipline attributes.Secondly, this paper employed the BERTopic model to extract the interdisciplinary internal topics and analyzed the evolution of internal topics.Finally, this paper conducted a comparative analysis of overall research and breakthrough research. [Result/Conclusion] This paper takes biomedical engineering research as an example, in terms of external discipline attributes, the degree of interdisciplinary is increasing, the key interdisciplinary sub-network of breakthrough research is a subset of overall research.The former pays more attention to"Medicine"while the latter pays more attention to"Engineering".In terms of internal topic evolution, the speed of interdisciplinary evolution is accelerating, the former is faster and the latter is relatively stable.The two have 5 and 3 promising topics that do not intersect.Based on the external discipline attributes and internal topic evolution, revealing the interdisciplinary development trend of overall and breakthrough research is helpful to promote interdisciplinary innovation.

[Purpose/Significance] This study aims to explore the influence of the interdisciplinary of paper on Disruptive innovation under different cooperation modes. [Method/Process] Taking the periodical papers in the field of bioinformatics from 2000 to 2018 from the core collection of Web of Science as data, this study quantified the interdisciplinarity of the paper from four dimensions: variety, balance, disparity and overall interdisciplinarity, and the Disruptive innovation with D index.Then it explored the influence of the interdisciplinarity of the paper on the Disruptive innovation under different cooperation modes by Logistic regression analysis, and tested the robustness of the conclusion by adjusting the citation time window. [Result/Conclusion] ① In the field of bioinformatics, the interdisciplinary of paper has a significant impact on Disruptive innovation: Variety and interdisciplinary have a positive impact on the output of Disruptive innovation, balance and disparity have a negative impact on the output of Disruptive innovation.②Under different cooperation modes, the direction of interdisciplinary influence on Disruptive innovation is basically the same, but the increase of cooperation span will strengthen the positive influence of variety on Disruptive innovation, and strengthen the negative influence of balance and disparity on Disruptive innovation, and weaken the positive influence of interdisciplinary on Disruptive innovation.

[Purpose/Significance] With the rapid development of health informatization, rural public health information services have become an important way to enhance the health of rural residents and improve their health literacy.The aim of this study is to reveal the evolutionary factors influencing the accessibility of rural public health information services under the background of digital healthcare, and explore their evolutionary mechanisms, in order to provide strong support for promoting the sustainable development of rural health. [Method/Process] This article focused on the accessibility of rural public health information services in the context of digital healthcare.A framework for analyzing accessibility of rural public health information services was constructed from three dimensions: pre-accesssibility, preliminary accessibility, and sustained accessibility, and research hypotheses were proposed.Samples were collected through questionnaire surveys, and empirical analysis was conducted by using binary logistic regression and ordered logistic regression methods. [Result/Conclusion] The research results show that digital access, digital capabilities, and service quality play an important role in the evolution process from pre accessibility to preliminary accessibility.However, the impact of health literacy and system quality on the evolution process from pre accessibility to preliminary accessibility is not significant.At the same time, perceived cost, social impact, information quality, structural assurance, and system quality play a crucial role in the evolution process from initial accessibility to sustained accessibility.The evolutionary mechanism proposed in this article during the process of"pre accessibility→preliminary accessibility→sustained accessibility", as well as the proposed countermeasures and suggestions, can help promote the accessibility construction of rural public health information services and enhance the overall health information literacy of rural residents.

[Purpose/Significance] Focusing on electronic medical record data to drive clinical decision making, the driving process model and the influencing factor model were constructed under the perspective of process theory and variable theory combination, which provided theoretical guidance for clinical decision support and promoted the value realization of electronic medical record data in clinical scenarios. [Methods/process] The driving constituent elements were analyzed, the driving process model was constructed, and the questionnaire survey and structural equation model were used to construct the influencing factor model based on clarifying the driving process. [Results/conclusion] Its core components include human beings, electronic medical record data, technology, clinical decision information, and organization and environment.The driving process follows the four steps of electronic medical record data generation and collection, data preprocessing and storage, modeling and analysis, and visualization and interpretation.Organizational environment, performance expectancy, and data quality significantly and positively influence the behavioral intention to use.Technology quality and data quality significantly and positively influence performance expectancy.The driving process model and the influencing factor model provided different but complementary perspectives to address the issue of how to drive.