[Purpose/Significance] In order to play the greater value of information, information science and information agencies, and promote the transformation of information research into think tank research. [Method/Process] This study used the methods of literature research and comparative study to define the concepts of information and think tanks, and analyze the relationship between them. [Result/Conclusion] This paper puts forward the "SQTSP Process" that information runs through the whole process of think tank research, and summarizes the keys to the transformation of information research to think tank research from five aspects: way of thinking, market nature, user-oriented, attribute characteristics and guiding factors, and three construction paths are proposed for the construction of information-supported think tanks.
[Purpose/Significance] This study automatically recognizes entities in high-quality medical knowledge with small samples, lay the foundation for the efficient and accurate processing of the structure of medical data in the form of text with small samples, lay the foundation for the efficient and accurate processing of the structure of medical data in the form of text with small samples. [Methods/Process] This study condensed the core concepts based on the concept system of lung cancer, and built a domain dictionary based on the core concepts.After segmenting and tagging the standard of diagnosis and treatment with this dictionary, the domain dictionary was integrated with the character level NER model Soft-Lexicon by using the character position label to construct an adaptive embedding method to form the Fusion-Lexicon model, and finally realize the NER of the corpus. [Result/Conclusions] The values of P, R and F1 of Fusion-Lexicon are 93.75%, 94.02% and 93.88% respectively, and its values of P, R and F1 of Fusion-Lexicon are about 4% higher than those of BiLSTM-CRF.The research is feasible and advantageous in realizing the ordering and structuring of high-quality medical knowledge with a small number of samples.
[Purpose/Significance] To provide the solution for the effective governance and application of intellectual property data resources, and promote the digital development of intellectual property and the upgrading of information public service system. [Method/Process] Based on the systematic investigation of the existing intellectual property public service platforms, the current situation of the organization and utilization of intellectual property public service data resources was analyzed.On the basis of the analysis of the basic structure and scene support of the data, the semantic organization connotation of the intellectual property public service data and its implementation path were further clarified, and the application effect was demonstrated by examples. [Results/Conclusion] The semantic organization framework from original data set to semantic enhanced data set is designed for public service scenarios, and the core technologies and practical principles for realizing semantic organization of intellectual property public service data are summarized to form a fine-grained data semantic organization and application pattern supporting multi-service scenarios.
[Purpose/Significance] The "cultural and museum promotion" has given rise to a large number of information resources related to cultural relics in the Internet, and how to extract entity relationships related to cultural relics is an important issue. [Method/Process] The paper proposed a new method of keyword extraction algorithm to extract content-related keywords from information resources as external knowledge base search terms, obtaining entity relationships from them and using distant supervision for entity relationship extraction of information resources after aligning multiple knowledge base entity relationships according to semantics. [Result/Conclusion] In the comparison tests with other methods, this method achieves better results under multiple keyword thresholds, integrates the advantageous features of multiple keyword extraction methods in obtaining the main contents of information resources, effectively solves the noise and relationship long-tail problems brought by distant supervision relationship extraction.
[Purpose/Significance] The purpose is to discuss the influencing factors of intelligent recommendation content"like"behavior in social media environment specifically, analyze its influencing mechanism and influencing path, aiming to provide reference for future research and practice. [Method/Process] The grounded theory was employed to conduct a three-level coding of interview data.Categories were sorted out and the relationship among them was analyzed.A theoretical model of the influencing mechanism of intelligent recommendation content"like"behavior in social media environment was developed. [Result/Conclusion] Platform reputation, intelligent recommendation information quality, intelligent recommendation service quality, intelligent recommendation system quality and user demand, user perception indirectly influence the"like"intention through user satisfaction.User perception, gamificated operations, social influence, conformity behavior directly affects the"like"intention, and the"like"intention directly affects the"like"behavior.Finally, suggestions are made from the perspectives of social media governance, operation and use.The future of research is prospected.
[Purpose/Significance] Augmented reality(AR)technology empowerment is a current research hotspot.Cognitive load is one of the important indicators to evaluate the efficacy of technology empowerment, but the impact of AR technology empowerment on users' cognitive load has not yet reached a unified conclusion.This study systematically analyzes the relationship between the two, in order to deepen the understanding of AR technology empowerment, to provide reference for the development of related industries. [Method/Process] This study included 15 experimental studies for meta-analysis.At the same time, the effects of moderator variables such as user education background, technology application field and experimental intervention time were analyzed. [Result/Conclusion] The main effect analysis shows that AR technology empowerment significantly reduces the cognitive load of users.Moderation analysis shows that technology application domains significantly affect the relationship between the two; higher education background users and short-term interventions have more significant effects.Therefore, in AR technology empowerment, it is necessary to combine user characteristics and content characteristics to effectively manage cognitive load to achieve the optimal effect of AR technology empowerment.
[Purpose/Significance] The health information of mobile medical services has the characteristics of privacy and sensitivity, and the disclosure of user health information has increasingly attracted the attention of the academic circles and the public. [Method/Process] Based on the elaboration likelihood model, this study used the methods of necessary condition analysis and qualitative comparative analysis of fuzzy sets to conduct empirical research on questionnaire data to explore the necessary conditions that affect the user's willingness for health information disclosure(WHID)and the configuration paths generated by different combinations of factors. [Result/Conclusion] The study finds that trust is a necessary condition to affect the user's WHID, and there are 3 paths to affect user's WHID: Path 1 is the risk resolution type, when the user perceives the risk to be eliminated, even if the information to be disclosed is highly sensitive, the user will have a high willingness to disclose; The Path 2 is the trade-off type, when the user's expected benefit is greater than the expected risk, his WHID will increase accordingly; Path 3 is profit-driven.when users' privacy concerns are low, their WHID is mainly driven by perceived benefits.
[Purpose/Significance] With the rapid increase of the amount of the academic information that users search and store, it has become a challenge for academic information searching.It is of great significance to study the obstacles and mechanism of personal academic information re-finding for solving the problems of personal academic information management, improving the efficiency of personal academic information utilization. [Method/Process] The Grounded Theory was used to analyze the obstacles to personal academic information re-finding, and the Interpretive Structural Models of personal academic information re-finding obstacles was constructed to analyze its' mechanism.Finally, the corresponding countermeasures were put forward. [Result/Conclusion] Memory lies on the surface of the obstacles of information re-finding, and losing of memory directly hinders the personal academic information re-finding; Ability, attitude, habits and information environment are located in the middle of the barrier factors which are easily affected by the underlying informational and environmental factors, and then affect the personal memory; Informational and environmental factors are the bottom obstacle factors of academic information re-finding, and they are the core and basic part of the whole mechanism.They affect the upper factors and thus act on personal academic information re-finding behavior.
[Purpose/Significance] The purpose of this paper is to discuss the influencing factors of users' knowledge innovation behavior in virtual communities.This will help people further understand users' motives behind their knowledge innovation and provide references and suggestions for improving users' knowledge innovation behavior in virtual communities. [Methodology/Process] From the perspective of sense of virtual community, intergrating the MOA theory and the S-O-R model, the paper the model of the influencing factors of users' knowledge innovation behavior in virtual communities and conducted an empirical test based on Smart PLS3. [Results/Conclusion] The analysis results show that the sense of membership, the influence and the sense of immersion have a positive influence on the will of knowledge innovation; group norms have a positive influence on knowledge innovation behavior and community trust has a negative influence on the will of knowledge innovation; innovation self-efficacy has a positive influence on the will of knowledge innovation; the will of knowledge innovation has a positive influence on knowledge innovation behavior.
[Purpose/Significance] User participation provides opportunities for enterprises to obtain heterogeneous knowledge resources and stimulate knowledge innovation.By exploring the knowledge conflict evolution process of user participation in innovation and proposing effective knowledge conflict management strategies, we provide theoretical guidance for enterprises to efficiently utilize user knowledge resources to form innovation advantages. [Method/Process] In the context of organizational user participation in innovation, the paper defined the connotation and types of knowledge conflict, explored the causes, development and results of different types of knowledge conflict, constructed the evolution mechanism of knowledge conflict in organizational user participation in innovation, and explained the"double-edged sword"effect of organizational user participation in innovation. [Result/Conclusion] With the knowledge exchange and interaction between organizational users and enterprises in the process of cooperation, knowledge conflicts stemming from the heterogeneity of inter-organizational knowledge resources are generated, and the evolution of different types of knowledge conflicts has a differential impact on innovation; content knowledge conflicts play a positive role in cooperative innovation, while schematic knowledge conflicts bring negative impact on cooperative innovation.Enterprises should improve contract content, enhance organizational flexibility, establish incentive systems that stimulate content knowledge conflicts, and promote cultural and value convergence to effectively improve the efficiency of collaborative innovation.
[Purpose/Significance] As a significant platform for Internet-based dynamic aggregation, organization and integration, and transformation and dissemination of evidence resources in social sciences, the evidence knowledge base in evidence-based social science can not only provide a best and reliable entry-point to access relevant research findings and evidence resources for researchers and decision makers, but also facilitate the use of relevant research results and evidence standards to quickly establish scientific basis and assist in formulating technology policies, which still play an important role in promoting evidence-based decision-making and practice. [Method/Process] This study conducted a survey of evidence portals with different topics in social sciences abroad, analyzed the relevant data with reference to the conceptual framework of existing evidence ecosystem and evidence standards, compared and summarized the differences, advantages and disadvantages of different evidence portals in the type and nature of evidence resources, evidence standards and relevant research methods.Finally, it discussed and put forward some suggestions and strategies about making evidence claims, adoption and evaluation of evidence base questions, application of evidence synthesis methods, and development of evidence-based guidance and practical methods for building the China's evidence knowledge base in evidence-based social sciences. [Result/Conclusion] This study can provide reliable evidence sources of resources, theoretical basis and reference of evidence types and standards construction for establishing the China's evidence knowledge base tool platform in evidence-based social sciences based on knowledge sharing network in the near future.
[Purpose/Significance] Studying the structural characteristics of the strategic task layout of some foreign libraries can provide reference for the scientific formulation and high-quality development of library strategic tasks in China. [Method/Process] Based on the grounded theory method, with the help of NVivo software, library strategic task texts from 10 countries were randomly selected for qualitative analysis. [Result/Conclusion] The research finds that there is a "pyramid" type common structure model in the elaboration and description of personalized task content:the bottom is an extensible strategic task content element system framework; In the middle is the core category, optimization category and support category module extracted from the text content element function; At the top is the structure of iterative optimization adjustment mechanism to ensure the function realization of content elements.
[Purpose/Significance] A full play is to be given to the role of digital technology in amplifying, superimposing and multiplying economic development, and the concept of digital economy and its platform economy is to be blended into competitive intelligence services for small and micro enterprises(SMEs), so as to solve the asymmetry of information resources prevalent in SMEs. [Method/Procedure] With the concept of digital economy and Internet thinking, the construction and operation mechanism of competitive intelligence service model for SMEs was optimized, on the basis of digitization, networking, and platformization, from the perspective of service structure, business structure, functional structure, technical structure, and business operation and data operation. [Result/Conclusion] The construction of the competitive intelligence service model and the optimization of its operating mechanism is aimed to provide solutions to competitive intelligence services for SMEs and help them to achieve high-quality and sustainable development.
[Purpose/Significance] The purpose of this paper is to construct the spatio-temporal evolution model and algorithm of online public opinion on public emergencies, revealing and presenting the spatio-temporal evolution characteristics and laws of public opinion from the spatio-temporal data obtained from the online public opinion on public emergencies, so as to provide certain guidance for the precise governance of public opinion on a spatio-temporal perspective. [Method/Process] In the construction of the spatio-temporal evolution analysis model of online public opinion on public emergencies, the global spatial autocorrelation in the spatial autocorrelation algorithm was used to analyze the overall spatio-temporal distribution of online public opinion, and the local spatial autocorrelation was used to analyze the types of spatio-temporal aggregation.The average clustering coefficient in the spatio-temporal network graph algorithm was used to analyze the network structure, and the local clustering coefficient was used to analyze key regions or cities. [Results/Conclusion] The spatio-temporal distribution of online public opinion on public emergencies has regional differences and certain scale characteristics, showing significant clustering characteristics in key urban agglomeration areas.The dissemination and diffusion of online public opinion event information has the effect of space-time contraction, and the spatio-temporal network of events presents a star-shaped and mesh-like information chain structure with individual cities as the core.
[Purpose/Significance] The phenomenon of disinformation reversal on social networks occurs frequently.Studying the reversal law and characteristics can effectively avoid the social disorder caused by the runaway evolution of public opinion on social networks. [Method/Process] Combining the disease dynamics model and the opinion dynamics model, the SIR-HK social network disinformation reversal model was constructed considering competitive propagation, and the empirical data of the "college entrance examination question card suspected of being swapped" event was collected to verify the effectiveness of the model.Besides, the law of disinformation reversal in social networks was revealed through numerical simulation. [Result/Conclusion] The appearance of disinformation will expand the scale of infected people, prolong the relaxation time of information transmission, and lead to the reversal of views spread by infected people.Moreover, the occurrence time of truth information will affect the number of infected people and the process of opinion reversal.What's more, the information with low credibility will affect the trust of the subject to the subsequent information.
[Purpose/Significance] The applications of digital government promote the linking, opening and value mining of public data, yet the continuous integration of public data often leads to personal information security issues.How to establish a personal information protection mechanism that adapts to the fusion of public data has become an unavoidable and important issue in digital governance research. [Method/Process] Based on the classification and life cycle theory of public data, this paper constructed an integrated analysis framework, namely "Data Risk Surface-Data Responsibility Line", and discussed the risk evolution of personal information in public data by analyzing multiple scenarios of Hangzhou City Brain. [Result/Conclusion] The case analysis shows that data sharing, data mining and data labeling are the main reasons that lead to the risk transition of private information in the public data, and behavior regulations for the government sectors would be further proposed such as department traceability, such as department traceability, prudent authorization, prior notification and ethical review, which are feasible paths for personal information protection in the context of digital government.
[Purpose/Significance] Through a review of the structural framework and content of privacy policies of representative foreign digital humanities crowdsourcing projects, this paper provides reference to the protection of personal information in digital humanities crowdsourcing projects in China. [Method/Process] This paper used Grounded Theory and LDA model to identify and code the privacy policies of 47 representative foreign digital humanities crowdsourcing projects, sort out the textual framework of privacy policies, and compare the structural framework and content of privacy policies of digital humanities crowdsourcing projects in different initiating platforms, countries and different types; Meanwhile, The current status of privacy policies of representative digital humanities crowdsourcing projects in China were also explored. [Result/Conclusion] The results show that the privacy policy includes seven dimensions:information content, information security, information storage, information sharing, information use, information collection and user power.The attention to the dimensions of Information Collection, Information Security and Information Content is relatively high, the attention to the dimension of User Rights is relatively low.The content characteristics of different implementation subjects, countries and different project types are different; Archives and museums in the initiating institutions and the UK and New Zealand in the initiating countries have higher index integrity in the privacy policy of digital humanities crowdsourcing projects.The integrity of transcriptional Digital Humanities crowdsourcing index is relatively low; Compared with the domestic privacy policies of digital humanities crowdsourcing projects, the category system of foreign policies is more complete and the content coverage is wider, especially the group of minors under the age of 14.Finally, this paper puts forward some suggestions on the protection of personal information of digital humanities crowdsourcing project in China from four aspects:privacy policy framework, industry self-discipline, Differences in privacy policy texts, the power of the subject and juvenile protection measures.
