[Purpose/Significance] High-value patents are important carriers of scientific and technological innovation.Identifying high-value patents and mining technological evolution characteristics from multiple dimensions are of great significance for discovering technological innovation elements and innovation laws. [Method/Process] Taking the patent literature in the field of"brain-computer interface"as an example, this study designed a patent value evaluation indicator system from three aspects of technical value, patent rights and market prospects, and then built a high-value patent knowledge graph.On this basis, this study analyzed the evolution of technological innovation from the dimensions of essence, application field, method, etc. [Result/Conclusion] This study finds that the evolution analysis method of high-value patent technology innovation based on knowledge graph has more diverse perspectives and deeper analysis levels, which can provide more specific reference for technology developers, technology policy and strategy makers to implement further innovation.

[Purpose/Significance] New word discovery, as the basic research of Chinese word segmentation, is very important for policy text mining, analysis and knowledge discovery.This paper intends to optimize the new word discovery algorithm and improve word segmentation performance in the AI policies.And we construct a domain lexicon to support the evolution study of AI policies. [Method/Process] A multi-feature fusion new word discovery algorithm(MFF)was proposed, which realized the mining of new word in the field of AI policies, and analyzed the evolution, continuation and diffusion of AI policies from the perspective of new word. [Result/Conclusion] The experimental results demonstrate the multi- feature fusion new word discovery algorithm effectively improves word segmentation performance and enriches the domain lexicon.The temporal changes in the emergence of AI policies neologisms reflect the key development areas of policy attention at different stages, and reveal the characteristics and evolution process of central and local governments in terms of policy innovation, continuation, diffusion and evolution.

[Purpose/Significance] Data reuse is a significant goal of opening research data.Current studies pay little attention on internal groups' characters within specific disciplines, especially in those matured in opening research data.Through identifying the demands of researchers data practice in data reuse activities, this paper conducts personas analysis to help optimize data services from information service institutions and effectively promote the development of research data reuse practice. [Method/Process] Following the logic of"demands identification-mechanism analysis-schema making", this paper conducted a questionnaire survey on the data reuse behavior of China's bioscience researchers.Through between-group differences analysis, the study analyzed respondents' attitudes and perceptions of data reuse among different data reuse practice states and conducted personas analysis of respondents by using hierarchical clustering analysis. [Results/Conclusion] The results show that China's bioscience researchers have a relatively low perception towards the foundation of community culture, perceived reuse regulation proficiency, and perceived reuse support usability in data practice.Based on the variables above, this study summarize two major types and six sub-persona types, which are"participants(including marginal type, expectant type, and preliminary exploratory type)"and"potential participants(including hesitant type, passive type, and wavering type)".Targeted and multi-layer service development suggestions are discussed, the study also proposes to develop progressive data reuse practice construction services, including immediate and multi-scenarios auxiliary services for research data reuse practice, short-term and multi-rounds' immersive experience services for research data reuse, and increasing mid-term to long-term project planning services focused on research data reuse memory points.

[Purpose/Significance] From the perspective of information chain, this paper explores the influencing factors and evolution rules of online health information quality governance users in different roles such as producers, disseminators and consumers, and proposes corresponding governance paths. [Method/Process] Firstly, the paper defined the different health information roles of users on the information chain and the costs and benefits of their corresponding health information autonomous behavior; Then, a tripartite evolutionary game model of"producer disseminator consumer"was constructed, and its dynamic evolution process and equilibrium stability strategy were analyzed; Finally, numerical simulations were conducted by using Matlab to explore the impact of parameter changes on user autonomy behavior and verify the accuracy of the model and evolution results. [Result/Conclusion] In the process of autonomy, the main factors that affect the game strategy of producers are"actively increasing"costs and trust benefits; The main factors driving communicators to shift from"conventional acceptance"to"active interruption"are feedback incentives and traffic returns; The main factors driving consumers to shift from"negative avoidance"to"positive avoidance"are feedback costs and positive incentives.

[Purpose/Significance] Timely and accurate identification of transformative research is of great significance for exploring the research frontiers in science, as well as facing the new pattern of high-quality development of the country. [Method/Process] Based on the characteristics of transformative research in the dimensions of breadth, intensity and speed, and by machine learning methods, the entropy weight method, factor analysis method and regression analysis method, the study constructed the transformative research identification model from the perspective of citation.Applied the models above to the fields of gene targets, magnetoresistive effects, fullerenes to evaluate the early recognition effect of different models, and applied the model with the highest early recognition effect to the fields of molecular biology, polymer science and astrophysics in 2017. [Results/Conclusions] Judging from the"ranking"and"top%"indicators, the recognition effect of the early recognition model of transformative research based on regression analysis is better than that of the other two models, and the identified transformative research ranks higher, and the transformative research could be identified in the third year after publication.The accuracy of the model in the field of gene targets and fullerenes are higher than in the field of magnetoresistive effect.The accuracy and recall rate of the early recognition model in the transformative study based on regression analysis are 80%, and the average score of the model is 0.764.

[Purpose/Signficance] The research on the scientific-technological-industrial interaction patterns based on the multi-relationship fusion, which is of great significance to reveal the related scientific base and the industrial development trend, aims to promote the industry-university-research deep integration. [Method/Process] Firstly, the topics of the papers, patents and products were identified by the methods of LDA models and human intelligence. Citation association, institution association, and content association between the papers and used the patents to evaluate the scientific linkage index. Then, the study used the application association between the products and the patents to evaluate the industrial linkage index. Finally, the empirical research was carried out in the field of quantum communication. [Result/Conclusion] The results show that the scientific-technological-industrial interaction modes in the field of quantum communication are the S pattern, the S-T pattern, the T-I pattern, and the S-T-I pattern. Among them, the S-T-I pattern increases synergistically with the passage of time; the technological evolution path in the field of quantum communication is "material-application model-key application technology-application system".

[Purpose/Significance] Studying how dynamic capabilities empower the knowledge creation of Open Government Data (OGD) would help to deeply understand the process of its dynamic capabilities and continue to promote the development of OGD. [Method/Process] The paper combined the theory of Complex Adaptive System (CAS) with the SECI model of knowledge creation, decomposed the process of OGD dynamic capability into three links of "detection-rule-response", constructed a functioning model of the OGD dynamic capability process based on CAS-SECI, deeply analyzed the knowledge creation core of the OGD dynamic capability through focus group interviews, and decomposed the OGD dynamic capability process. [Result/Conclusion] In the process of "detection-rule-response", the change perception ability of OGD, the absorption and transformation ability of OGD, the communication ability of OGD and the reconstruction and innovation ability of OGD realize the transformation of knowledge capital through the relevant knowledge activities and processes in knowledge socialization, externalization, combination and internalization, which systematically reflect the knowledge creation process of OGD's dynamic ability. Focusing on the implementation and optimization of related knowledge activities, the paper puts forward countermeasures and suggestions for strengthening detection, focusing on accumulation, strengthening communication, and focusing on innovation.

[Purpose/Meaning] The government is the main body of data governance in the whole society, and evaluating the effectiveness of government open data is conducive to building a new model of government data governance in the digital era and promoting the modernization of governance systems and governance capabilities. [Method/Process] Starting from the two dimensions of management system and value system, the evaluation index system was constructed. At first, the entropy weight method was used calculate the weights of each index, and the TOPSIS method to evaluate the comprehensive score of the data governance efficiency of the sample cities, then the study used the rank and ratio method to rank the comprehensive score, and finally used the correlation analysis to find the key elements. [Results/Conclusions] The results show that the efficiency of government open data governance in Shanghai, Beijing and Guiyang is excellent. The efficiency of government open data governance in Wuhan, Guangzhou, Chengdu and other cities is mid-range; Ningbo's government open data governance efficiency is average. Moreover, the correlation between the development level of digital government and the effectiveness of government open data governance is the highest.

[Purpose/Significance] Information collaboration state is an important factor affecting the level of smart government construction. At present, the level of smart government construction in various cities in China is uneven. By constructing a supernetwork model of smart government information collaboration, the information collaboration state is quantitatively measured, and the research on the division of inter-city information collaboration is conducted based on real city cases. It is of great significance to clarify the goal and direction of the information collaborative construction of smart government affairs and promote the development of smart government affairs. [Method/Process] This paper constructed a supernetwork model of smart government affairs, proposed a quantitative measurement scheme of information collaboration state from micro and macro perspectives, selected representative cities in different stages of smart government affairs construction in China as cases, deeply analyzed inter-city differences of various measurement indicators, and revealed the status quo of inter-city division of information collaboration construction according to the mapping of indicators to practical significance. [Result/Conclusion] The analysis finds that with the improvement of the level of intelligent construction, the overall entropy of information collaborative network shows a trend of increasing, that is the correlation between information became more complex, and the scale of information collaboration would continue to expand. But the speed of enriching links between nodes is usually slower than the expansion of the scale of collaboration, that is, the development speed of the correlation at the information content level is lower than the growth speed of the demand for information collaboration. Based on the characteristics of information collaborative networks, this paper puts forward the corresponding information collaborative development strategy from the single element and the whole macro level.

[Purpose/Significance] Paper classification is a core issue in scientometrics, holding vital implications for practical research assessment. [Method/Process] This study introduced an innovative approach to classifying individual papers, leveraging multigenerational references. The core idea involved extracting information from the multigenerational references of the target paper to construct a more comprehensive disciplinary structure, subsequently assigning the target paper in one to three disciplinary categories. To validate the efficacy of this method, the study employed papers indexed in the Web of Science Core Collection between 1999 and 2018 as its research subjects. The investigation compared the disciplinary classification results of individual papers and explored application scenarios from the perspectives of disciplines and journals. [Result/Conclusion] The findings suggest that incorporating disciplinary classification information from multigenerational references reduces the entropy of disciplinary classification in the target paper, significantly enhancing the accuracy of disciplinary classification for individual papers. Through the application of diverse weight-setting rules and strategies for handling multidisciplinary science, this method, to a considerable extent, tackles disciplinary classification challenges present in literature from multidisciplinary journals. Furthermore, it offers a practical solution for identifying interdisciplinary papers.

[Purpose/Significance] The study aims to analyze the impact of academic background differences on interdisciplinary research output core authors in the field of library and information: to provide valuable insights for formulating policies on the management and evaluation of scientific researchers, as well as for fostering high-level interdisciplinary research teams. [Method/Process] Firstly, a high-output scholar group was identified by analyzing publication records in core journals, from which core scholars were selected as the target scholars. Secondly, the resume information of the target scholars was collected and encoded to capture academic background differences. The research output in terms of interdisciplinary publications within five years of obtaining the highest degree was collected to measure the interdisciplinary nature of their academic work. The mean score was calculated to obtain an indicator of scholars' interdisciplinarity. Finally, correlation analysis and multiple regression analysis were conducted to investigate the relationship between academic background differences and scholars' level of interdisciplinarity. [Results/Conclusion] Results indicate that younger scholars with more recent graduation dates and differences in academic and professional development backgrounds are more likely to engage in interdisciplinary research. Among these factors, differences in academic development backgrounds can encourage scholars to engage in interdisciplinary research, but do not directly affect their degree of interdisciplinary output. Differences in graduation dates and professional development backgrounds can have an impact on the level of interdisciplinary output in scholars' research papers.

[Purpose/Significance] The innovation of academic paper is not only a result, but also a process.The stage division of innovation process of academic paper is helpful to clarify the measurement angle of academic paper in each stage and the current research progress. [Method/Process] Based on the discussion and interpretation of the related concepts, as well as the logic and timeliness of the innovation process, the innovation process of academic paper was divided into the stage of innovative knowledge absorption, the stage of innovative knowledge output and the stage of innovative knowledge diffusion.Then, this article systematically combed and summarized the innovative measurement indicators, measurement angles and main research conclusions of each stage, and summarized the realistic difficulties of existing research. [Results/Conclusion] In view of the specific research problems of the innovation of single academic paper, fine-grained analysis should be carried out from different stages, so as to draw scientific conclusions and put forward research prospects for subsequent empirical work, in order to provide reference for the research on the innovation measurement of single academic paper.

[Purpose/Significance] This paper summarizes the research progress of health information adoption behavior at home and abroad, finds out the achievements of existing research, and puts forward potential research directions worthy of attention in the future, so as to provide reference for the follow-up theoretical study. [Method/Process] By systematically combing the collected domestic and foreign literature on health information adoption behavior, this paper defined the concept of health information adoption behavior, summarized the manifestations of health information adoption behavior, and summed up the development and evolution of the theoretical model of health information adoption behavior. [Result/Conclusion] In the future, further research on health information adoption behavior can be considered from four aspects: the composition of prefactors for health information adoption behavior, the health information adoption behavior under context nesting, the mediating effect of health information adoption execution intention, and the measurement of actual health information adoption behavior.

[Purpose/Significance] An in-depth analysis of the research hotspots and frontiers of global health information adoption behavior, aiming to show the research progress in this field and provide research directions and methodological references. [Method/Process] Tools such as VOSviewer were used to visualise and analyse the issuance of existing research results, in order to reveal the development trends and research hotspots in this field. [Result/Conclusion] The foreign works focuse on four themes: the behavioral mechanisms and influencing factors of health information adoption, as well as the service framework and public decision-making from the perspective of health information adoption behavior; the domestic literature focuses on three themes: study on the influencing factors of information adoption in online health communities, study on health communication from the perspective of health information adoption, and the predictive analysis of health information adoption behaviour.The research frontiers mainly focus on two aspects: model construction for health information adoption behaviour, and user characteristics that influence health information adoption behaviour.