［Purposeg/Significance］Innovation driven, intelligence first, in a wider range of strategic height, integration of social economy, scientific and technological development, horizontal analysis and other related intelligence, deep mining of information resources, need new information science theory, technology and method to support.In order to understand the current situation of information science research supporting innovation at home and abroad, reveal the hot topics and frontier trends in this field, and make some preparations for the construction of innovative information science.［Method/Process］Using bibliometric method, domestic CNKI and foreign web of science core databases were selected as data sources, and CiteSpace software was used to analyze the trend of published papers, keyword clustering and detection of emergent words.［Result/Conclusion］It is found that the research hotspots of Chinese and foreign journal papers were diversified, including patent intelligence analysis oriented to innovation, knowledge management oriented to innovation, and information science methods oriented to innovation identification;in future research, following topics should be focused on: evaluation of innovation oriented academic achievements, identification and extraction of innovation information, organization and retrieval of innovation information, and promote the development of information science research supporting innovation to the construction of innovation information science.

［Purposeg/Significance］Identifying, organizing, and searching for innovations in academic papers is not only conducive to improving the efficiency of scientific research and promoting knowledge discovery and innovation, but also helping to enrich the ways of innovation monitoring of scientific papers to achieve the purpose of promoting scientific research innovation.［Method/Process］This study takes information science journal papers as an example, uses deep learning methods to identify innovative sentences in academic papers, and realizes the multi-granularity association of the innovative content of academic papers.Based on the multi-granularity associated results, the innovation points retrieval system is constructed to provide the entrance of innovation points retrieval.［Result/Conclusion］The methods explored in this study can identify and organize more fine-grained paper innovation points, provide a certain foundation for innovation information retrieval, innovation knowledge discovery, and academic paper innovation evaluation, which contribute to scientific and technological innovation.

［Purpose/Significance］This study deepens the perspective of knowledge graph construction to the innovative content with core value within academic papers, carries out explorations in the construction and application of the knowledge graph of the innovative content of the papers, and exerts their effectiveness of academic papers from the perspective of innovation to improve its utilization efficiency.［Methodology/Process］The key object analysis, a pre-processing phase of knowledge graph construction is formed by combining word frequency analysis and semantic analysis methods such as mutual information and left-right information entropy, which provides an important reference for the initial construction of ontology, and at the same time combining ALBERT deep learning model for knowledge entity identification, SmartKG and MongoDB are used to build the knowledge graph and then its potential applications were discussed.［Results/Conclusions］The key object analysis phase adopted in this study can effectively mine the key information of specific domains, and the knowledge graph constructed in this study contributes to the revelation and utilization of innovative contents of academic papers and leaves convenient conditions for further knowledge fusion.

［Purpose/Significance］In order to explore the coping strategies and interaction of different subjects during the epidemic, this paper analyzes the multi-dimensional data of the government, media and people of China, the United States and the United Kingdom during the COVID-19.［Method/Process］The coping strategies of different agents were revealed by subject extraction and content analysis, and the evolution of coping strategies of government, media and people in epidemic was analyzed and summarized.［Result/Conclusion］According to the analysis results, the role of the three agents of response strategies(government, media and people)was further explored in response to public health emergencies, and how the response strategies among different subjects affected each other, so as to provide policy suggestions for improving the response mechanism of public health emergencies.

［Purpose/Significance］Scientific paper is a kind of argumentative text, which demonstrates scientific viewpoints and conclusions.The analysis and semantic representation of scientific paper's content from the perspective of scientific argumentation can more accurately represent the content of scientific papers and help improve scientific research personnel's quick reading and understanding of scientific papers.［Method/Process］Based on Popper's knowledge growth theory to explain the process of scientific research, seven kinds of argumentation elements in scientific papers were summarized: previous statement, problem, hypothesis, goal, method, result and conclusion.Based on Toulmin's argumentative, a sentence and entity-level scientific paper argumentation structure ontology was proposed.And on this basis, an argumentation structure ontology application prototype retrieval system was designed from two type of content-based semantic retrieval and knowledge visualization.［Result/Conclusion］The results showed that the argumentation structure ontology constructed in this paper can accurately represent the argumentation elements and their argumentation relationships in the scientific paper, realize semantic retrieval and knowledge discovery based on the content of the paper, and provide a reference for the research on semantic publishing of scientific paper.

［Purpose/Significance］The study explores why some questions are highly rated by emphasizing the knowledge characteristics embedded in the questions, so as to provide reference for optimizing the quality of user-generated content and realizing effective knowledge sharing.［Methods/Process］Data from a programming-related Q&A site—Stack Overflow for ten years were collected, spanning the years from 2008 to 2017.The influence of knowledge characteristics on question score was analyzed from four aspects: knowledge popularity, knowledge correlation, knowledge diversity and knowledge importance.［Results/Conclusions］The research found that knowledge characteristics had a significant impact on content quality.Questions with highly relevant tags and tags of high network importance were more likely to get high scores.The tag frequency was curvilinearly related to the likelihood of attention allocation to questions in an inverted U-shape, which showed that proper tag frequency helps to improve question evaluation.

［Purpose/Significance］In order to solve or improve the problems of personal privacy and security, ownership of data, data isomerization and other issues in the traditional scientific data sharing model, a personal scientific data sharing model based on blockchain was proposed, which is expected to promote the practical and theoretical research of personal scientific data sharing.［Method/Process］On the basis of systematically sorting out relevant research at home and abroad, the advantages of blockchain technology and SQL database were combined, the architecture model was constructed, and its process, characteristics and implementation technology were analyzed.［Results/Conclusion］The decentralized model based on blockchain technology has the advantages of high flexibility, strong security, traceability, and trust.It can effectively promote the sharing and circulation of personal scientific data, and then better play the value of scientific data.

［Purpose/Significance］In order to effectively alleviate the contradiction between the growing demand for emergency care knowledge of the public and the current insufficient supply of relevant public science resources, this paper aims to construct an emergency care knowledge service platform oriented to the public demand for popular science.［Method/Process］Firstly, the public demand for knowledge popularization of emergency care was summarized and analyzed.And then on this basis, an emergency care knowledge service platform system framework consisting of three levels was constructed: knowledge collection layer, knowledge organization layer, and knowledge service layer.In this paper the operation guarantee mechanism of the knowledge service platform was discussed from the perspectives of knowledge alliance, expert collaboration, user rewards and punishments, popular science rumor refutation, publicity and promotion.［Result/Conclusion］The platform can provide users with emergency care knowledge resources that are authoritative, accurate, timely updated, rich and diverse, and come out personalized and accurate knowledge service, so as to effectively meet the public demand for knowledge popularization.

［Purpose/Significance］As a new educational model, mobile learning has greatly improved the convenience of information resource access.The course knowledge payment willingness model of mobile learning platform built in this paper can provide operational path reference for platform operators to optimize user services.［Methods/Process］Drawing on the SOR model, the data was collected through the questionnaire survey to deconstruct the influencing factors related to the knowledge payment willingness of the course mobile learning platforms from the two aspects of social influence and content characteristics.Data analysis and hypothesis testing were performed on 378 valid sample data with the help of SPSS22.0 and AMOS22.0 data analysis software.［Results/Conclusion］The results showed that when users were stimulated by two external factors: social influence and content characteristics, they would produce a process of payment recognition of the content on the mobile learning platform, which would then affect their willingness to pay.In addition, expected revenue and personal payment awareness also had a positive impact on users' willingness to pay.

［Purpose/Significance］The introduction of the Personal Information Protection Law provides a detailed normative basis for the protection of readers' personal information, but it still needs in-depth analysis of the civil law protection mode for readers' personal information, so as to accurately achieve the balance between the protection and utilization of readers' personal information.［Method/Process］Through the whole legal order methodology, the reader's personal information was divided into reader's privacy information, reader's sensitive personal information, reader's general personal information and reader's borrowing information.Through the doctrinal methodology and from the perspective of scene theory, fair information practice theory and natural rights theory, the paper sought for the hierarchical and orderly protection of readers' personal information.［Result/Conclusion］Readers' privacy information should be protected in the mode of empowerment, and only express consent could be taken as the only legitimate use.For readers' sensitive personal information, the right mode of behavior regulation should be chosen and the opt-in mechanism referring to intellectual property protection should be adopted.For readers' general personal information behavior regulation mode should be chosen to balance the multiple interests with opt-out mechanism and economic incentive mode.The protection mode of readers' borrowing information is not independent.It is necessary to distinguish specific situations and apply the aforementioned three protection modes or data protection modes.

［Purpose/Significance］The guidance of network public opinion is an important force for the harmonious ecological environment of network public opinion.The multi-subjects that guide online public opinion are interdependent and co-evolve.There are multilateral and multi-directional communication mechanisms and symbiosis models that together form a complex symbiotic ecosystem.Research provides guidance for the formation of healthy and coordinated symbiotic relationships among multiple subjects.［Method/Process］Three main subjects were taken as the entry point, closely followed the symbiosis theory of population ecology, a network public opinion multi-subject-guided symbiosis model was built based on the logistic model, the balance point and stable conditions of the model was analyzed, and numerical simulation was used to explain 7 types of the law of symbiosis evolution between subjects under the symbiosis mode, and the influence of different initial values on its evolution trend.［Results/Conclusions］The results showed that symmetric mutualism was the best mode of multi-agent cooperation, and the guiding agent should be committed to promoting the transformation from parasitism and profit oriented symbiosis to symmetric mutualism.The guiding effectiveness in a stable state was positively correlated with the symbiosis coefficient and the maximum guiding effectiveness and its growth rate was negatively correlated with the initial guiding efficiency, and positively correlated with the inherent growth rate.

［Purpose/Significance］The objective weighting methods are widely used in the evaluation of academic journals.The principles of different objective weighting methods and their comparisons need to be further studied.［Method/Process］Based on the journals of mathematics, economics and computer science of JCR2019, weights, evaluation values and evaluation orders of different objective weighting methods were compared.［Result/Conclusion］The research showed that the principle of objective weighting methods mainly included three aspects;the differences of objective weights were great but the difference of evaluation result was not always great.The weights and evaluation results of objective weighting method with similar principles were highly correlated.The objective weights were often inconsistent with the subjective management needs.In the actual journal evaluation, whether the weight obtained by objective weighting method meets the needs of scientific research management should be considered.

［Purpose/Significance］Based on the perspective of citation time, the article explores the relationship between the internal and external characteristics of highly cited papers and the citations.The specific research questions include three aspects: (1)the relationship between the internal and external characteristics of highly cited papers and their initial and long-term citations;(2)The relationship between the internal and external characteristics of highly cited papers and citations over the years;(3)The relationship between the internal and external characteristics of highly cited papers and the citation speed.The research results have reference value for the prediction of highly cited papers.［Method/Process］Based on the over-dispersion test of the citations of highly cited papers, the relationship between the internal and external characteristics of highly cited papers and their citations was expbred by negative binomial regression and multiple linear regression.［Results/Conclusions］The result showed that internal and external characteristics of highly cited papers jointly affected the citations, and their influence on the citations also showed certain differences: internal and external characteristics of highly cited papers had a greater impact on initial citations than long-term citations;characteristic variables such as the scale of national cooperation, the number of disciplines, funding, journal impact factors, and open access always had a greater impact on citations, which presented certain regularity;characteristic variables such as long title, multinational cooperation, funding, high impact factor journals, and open access had a great effect on the citation speed.

［Propose/Significance］SciVal is an emerging and comprehensive analysis platform/tool for accessing to research performance and disciplinary competitiveness.Studying and exploring the SciVal is of significance for the evaluation of research performance and the planning of disciplinary development.［Method/Process］In this paper, SciVal was applied to evaluate the research achievements in the field of library and information sciences of 26 Chinese universities, from the aspects of overall research output, research topics, collaboration status, author contributions and social impact.Then, the evaluation of Chinese universities was further compared with some top universities in Europe and USA.［Result/Conclusion］The results showed that overall research output, research topics, and collaboration status of top Chinese universities were comparable to, or even exceed some top universities in Europe and USA.But the competitiveness and impact of research achievements in top Chinese universities have a certain gap with some universities in Europe and USA.

［Purpose/Significance］The legal benefit of private information in the era of big data has become a research hotspot at home and abroad.Scientific mining and revealing the knowledge structure and cutting-edge trends in this field can provide references for subsequent research.［Method/Process］Based on existing litterature, the relationship model, document structure, historical evolution and cutting-edge topics of foreign research knowledge of the legal benefits of private information were visually analysed.［Result/Conclusion］Firstly, foreign research knowledge has experienced five stages, which are the budding of traditional information, the start-up of mainframes, the understanding of the Internet, the deepening of social media, and the expansion of big data.Secondly, the development of personal solitude and four types of privacy infringements, personal information personality and property rights control, privacy culture and social norms & other legal interests, personal information identifiable, anonymized and re-identified, and five research lines for the legitimacy of the use and protection of private information was presented.Finally, it coverd three levels of "legacy of traditional privacy information, legal benefits of new privacy information and the legitimacy of privacy information control" and nine clusters of "privacy, personal information, data protection, personal data, surveillance, consent and confidentiality".It mainly involved frontier hotspots such as self- determination control, property right confirmation, balance of legal interests, reasonable expectations, risk assessment, informed consent, and insensitivity to violations.

［Purpose/Significance］To make a comprehensive and systematic review of academic resource mining methods, the paper compares and analyzes different mining methods, and discusses their application and future development direction.［Method/Process］Based on the search results of literature database on academic resources mining, research topics of these articles was analysed, mining methods from perspectives of research object, mining dimension and mining technology were compared, existing mining methods were divided into two dimensions: vertical mining and horizontal mining, and then the application in academic search and recommendation, research and technology frontier identification and prediction was discussed.［Result/Conclusion］At present, the main research objects of academic resource mining are still academic papers and patents, which need to strengthen the comprehensive mining and cross-domain mining of academic resources, while the application research of knowledge graph, deep learning, big data and other technologies in academic resource mining needs to be further broken through.