[Purpose/Significance] There are complex network structures in the internal session process of WeChat Group and it is of great significance to explore the characteristics of the internal network of WeChat group and to divide the members' roles in the study of the relationship between the information dissemination and the member roles in the WeChat group.[Method/Process] This paper was based on the theory of virtual community and the theory of session analysis,and the conversation data within the real WeChat group was taken as the research object,by constructing membership type of WeChat group members,the membership intensity algorithm was designed based on commom response and mutual response of members,and the conversation exchange network was constructed.Social network analysis and content analysis were used to analyze its static structure and dynamic structure evolution analysis.[Result/Conclusion] The roles and positions of members in WeChat group were different,and the process of conversation exchange showed a clear central tendency.The WeChat group's conversation exchange network conformed to the six-degree division theory and had the characteristics of small-world effect.The WeChat group's conversation exchange network had the characteristics of dynamic evolution core members were time-dependent.

Social search is a search model based on web2.0,which is newly emerging and attracts people's attention rapidly.Medical social search,based on social search model,has been a prevalent way for people to search medical information and ease the tension of medical resources effectively.This paper constructed the user-centered satisfaction model of the search results of medical social search platform based on the health belief model and customer satisfaction theory,and verified and revised conceptual model by survey data.The results showed that:The sources quality was the key quality indexes influencing user's health behavior expectation and perceived health threat;the emotion quality predicted user's self-efficacy;the user's health behavior expectation and perceived health threat had a more significant influence on user's perceived value than user's satisfaction,while user's perceived value and self-efficacy was the key indexes influencing user's satisfaction.

In order to advance the public participation and enhance the level and depth of public participation in smart city construction and management,the paper proposed a crowdsourcing framework on smart city management to public participation in social media environment based on the theories and practices of public participation and crowdsoucing.The proposed framework consisted of government sector,crowdsourcing project operator,social media platforms and publics,and which had good practicality and operability built on the social media platform.Finally,the framework was explained.

The paper analyzed the process and influencing factors of new think tank product dissemination,improved the dissemination efficiency and influence of think tank products,and provided theoretical basis and reference suggestion for the promotion of new think tank products of new media platform.Theoretical model of influencing factors of new think tank product dissemination under the new media environment was proposed based on the theory of planned behavior.The data were collected by questionnaire,and the partial least square structural equation modeling was used to test the hypotheses.In the new media environment the factors of think tank products,think tank product dissemination subject of think tank and think tank users,the environment of think tank positively impacted on the think tank product dissemination intention,the factors of think tank product dissemination technology directly affected the think tank product dissemination behavior.

The ecological efficiency of China's export products is generally low,which is easy to be charged for ecological dumping by various countries in the world.Establishing a synergetic competitive intelligence system with enterprises as the main body,industry associations as support,government-led,information service agencies as intermediaries,collecting,processing,arranging,analyzing and researching the relevant information at home and abroad,was an effective way to deal with foreign ecological anti-dumping.The construction index for competitive intelligence system to deal with the ecological anti-dumping,the application of the grey clustering evaluation model with the confidence degree,was beneficial to continuously improve the present situation of the competitive intelligence system,and improved the ability of Chinese export enterprises to cope with the ecological anti-dumping.

[Purpose/Significance] Information design aims at presenting the complex information in a clear,fast and comprehensible way,which provides a new perspective for intelligence research service innovation.[Method/Process] Based on information design theory and intelligence service theory,the paper explored the necessity and feasibility of expounding intelligence service mode of combining information design,the connotation,elements,service flow of new mode.[Result/Conclusion] This paper put forward the realization path from four aspects for providing theoretical and practical reference for intelligence service innovation:enhancing information design consciousness for serving users' decision-making needs,strengthening information design literacy for improving information design ability,adjusting organization operation mode for promoting the integration of research and design,expanding training channels for enhancing efficiency of information design.

[Purpose/Significance] As a product of "Internet+government services",integrated functions,numerous users,vast amounts of information and multiple demands of Government Affairs Weibo provide a new perspective for social governance.[Method/Process] The 2000 Government Affairs Weibo randomly selected from the database was used as research objects.The functional mechanism of the government weibo in social governance was analyzed through SPSS19.0;by using the case analysis method,the@Central Meteorological Observatory and@Asking Political Yinchuan as well as@Beijing weibo Group were selected to empirically study content-based visualization,efficiency values based on timeline,and weibo group effects based on emergencies in the unique mode of social governance.[Result/Conclusion] The results showed that Government Affairs Weibo had a significant role in predicting Internet public opinion and innovating social governance.The good idea could contribute to multi-planning(co-construction)and wide participation(co-governance)in the context of information exchange,sharing(sharing)of the socialization mechanism, in order to provide a certain reference for the government to better use the new media "listen to public opinion,benefit people's livelihood,solve people's concerns".

[Purpose/Significance] Most of the social network node's influence calculation does not consider the user's evaluation,and the user's evaluation has important significance to the identification of the professional influence nodes in a specific field.[Method/Process] In this paper,a domain dictionary and a topic recognition model were used to limit the subject range of the target users.At the same time,the link network was established based on the user concerns,and the sentiment score of the user reviews was fully incorporated into the Domain Rank calculation for the mining of professional influence nodes.[Result/Conclusion] Our study showed that the algorithm could effectively find and identify potential domain influence nodes from multi-topic user groups.

The wide application of WeChat moments makes its influence improved greatly,and the information dissemination of WeChat moments becomes a research hotspot.In this paper,the information dissemination mechanism of WeChat moments was analyzed,five characteristics were condensed from the users,users relationships and the information contents,the prediction method of the sharing model was established based on the LR algorithm,and then the information dissemination model of WeChat moments was constructed,the model was validated by the experimental data that obtained from the research,which was very important to reveal the mechanism of the information dissemination.

Tag-based personalized recommendation is becoming more and more popular,but the semantic ambiguity and temporal dynamics of the labels affect the quality of personalized recommendations,and the existing researches only consider the relationship between users and labels in terms of quantity and structure.Personalized recommendation based on social tagging system firstly mined potential semantic topics of the tags that fused social relationships,then built multi-level,multi-dimensional user interest models,and proposed a model update strategy,finally achieved personalized recommendations.Collecting CiteUlike site data for experimental analysis,and the results showed that the improved algorithm expressed user interest preferences more accurately than traditional algorithms,and effectively improved the accuracy of personalized recommendation.

The paper aimed to achieve accurate information recommendation for subject groups,meeting the demand of researchers to obtain information more conveniently and quickly.First,it gathered knowledge resources to build a subject groups platform knowledge circle.Then,it constructed a subject groups interest model and calculated similarity between recommendation information and subject groups interest.Automatic and accurate recommendation of the scientific and technological information and the field monitoring express was realized on Subject Groups Integration Platform,which supported sustainable development of subject groups Integration platform.

[Purpose/Significance] The function of checking and novelty searching is one of the information services provided by digital libraries,and plays an important role in the healthy development of academic circles.However,most of the current detection system detects the text,the detection and recognition function of image has great limitations,which lead to some authors in the detection process opportunistic.Therefore,it is of great significance to study image false detection.[Method/Process] In view of this problem,this paper proposed a model that can identify the image forgery in the paper,and designed the frame of the image data detection platform.[Result/Conclusion] The experiment proved the feasibility and validity of the model.It could provide reference for image tampering detection of University Library and improve the quality of information service,and had certain theoretical value.

[Purpose/Significance] The paper proposed the concept of keyword set,explored R language programming to achieve the creation and visualization of keyword set co-occurrence network,for further research on keyword set-based data mining and knowledge discovery to provide a more general way and method.[Methods/Process] Using R language programming technology and igraph contribution package,the paper designed the key word set co-occurrence network creation and visualization function,and analyzed the published data of 18 CSSCI source journals in the field of graph science.[Results/Conclusion] The centrality and other characteristic parameters of the co-occurrence network of keyword set were calculated,and the co-occurrence network graph of keyword set was drawn.The research showed that the co-occurrence network of keyword sets revealed the distribution,clustering and relationship characteristics of keyword sets,keyword sets could more intuitively reveal the theme content and its relationship of the analysis object,and the construction and characterization of its characteristic parameters were worthy of systematic and in-depth study.

With the demand change from the preservation of resources to the using of resources,after years of development,institutional repository gradually from the traditional collection of resources to knowledge associated and discovery.How to deep integration with data association management becomes particularly important.From the prespective of science research support and teaching support,the article looked for the relationship of all kinds of resources between,resources and persons,resources and organizations,resources and groups,attempted to build institutional repository data model,look forward to guide the deep and multidimensional linked data during the construction of institutional repository.

Under the background of "double first-class" Construction,there is less research on the subject-precision service of academic library,in order to serve the traditional subject,it is necessary to strengthen the research of precision service in the subject construction of university library,which is of great significance to the development of subject service under the background of "double class".This paper investigated and analyzed 42 "Double first-class" University library subject service,sumed up the current situation of subject service in universities,and put forward the countermeasures for the accurate service of library subject from 4 aspects,so that the library resources service and intelligence service function were optimized,which aimed to better support the construction of subject service in order to promote the subject service of university library to change rapidly to the subject precision service direction.

[Purpose/Significance] The main research methods and open source tools used in path recognition research of scientific and technological innovation are analyzed.[Method/Process] Firstly,from different angles of citation analysis,structure analysis,hole subgroup analysis,text mining,content mining based on Citation Method of science and technology innovation path recognition analysis;secondly,for the 3 open source tools of technological innovation path to realize the recognition of citation content mining from the data source,and text format mining content,processing capacity and other aspects:Mallet,Knime,NLTK;finally,it analyzed the realization of the 5 open source tool path visualization technology innovation:Pajek,Ucinet,Gephi,D3.js,VOSViewer.[Result/Conclusion] through research,it is found that using open source tools to mining rich semantic information contained in citation contents,identify different citation motivations and visualize display can effectively achieve the expression of innovation path.

In the E-Science environment,research activities are characterized by data intensive.Data management has gradually become a hot topic of common concern in the scientific community.Six University Libraries in the United States carry out research data management services from the establishment of special service teams,education training and consulting services,and the construction of technical platform.Through the network research method and literature analysis method,this paper worked out the successful experience of data management service,and summarized several enlightenments for domestic university libraries.Our country should draw on foreign advanced experience,formulate relevant national policies,construct communication and cooperation mechanism of relevant institutions,build the infrastructure of data storage,and improve service ability.

The paper built an institutional knowledge base platform integrated into the scientific research management process,and formed a new institutional knowledge base platform consisting of the research project management system,the scientific research personnel management system,the scientific research achievement management system and the scientific knowledge reward management system,which was to achieve institutional inheritance and resource traces of knowledge assets.For the new knowledge base platform built,this paper proposed the resource acquisition process and proposed six strategies:strengthening management awareness,standardizing resource construction standards,emphasizing trace management of outcomes,strengthening the knowledge organization system,attaching importance to service functions,and establishing rights protection mechanisms in order to jointly promote the establishment of an institutional knowledge base integrated into the scientific research management process.

Based on WoS,the paper collected a total of 4023 research papers related to health information.By selecting keywords as the units of analysis,co-word analysis and cluster analysis as methods,SciMAT as the knowledge mapping tool,this paper identified research topics and their evolution paths in different periods.Research demonstrated that the study of international health information began in 1923,with slow development,a few research topics and less attention in the early stage.Then the research turned to diversification.Research topics changed greatly in the different periods and evolutionary relationships were complex.Research also revealed four evolution paths on two different research directions.Evolution paths of health information seeking and health literacy were important and would continue to be hot topics.Evolution paths of public health and health information technology received less attention with unstable evolutionary processes in the early stage.After 2008,topics of the two paths received widespread attention and high research status,which became research hotspots and the basic research direction with development potential.

[Purpose/Significant] Whether the technology similarity between enterprises can be transformed into the real collaboration relationship? Whether the potential correlation disclosed by patent coupling indicates collaboration or competition? Answers for these questions must be deduced from empirical research.[Method/Process] With the methods of visualization and factor analysis,this paper constructed patent bibliographic coupling network based on Derwent patent classified codes to investigate enterprises' technology similarity and cluster structure,and explored technology cliques and R&D directions,as well as compared the patent bibliographic coupling network and patent co-invent network.[Result/Conclusion] The simple technology similarity was not the effective basis to identify the potential collaboration relationship.The potential collaboration relationship based on patent bibliographic network was hardly to transform into reality.Social relation factors were more significant than technology similarity to determine the patent collaboration of enterprises.

[Purpose/Significance] For the purpose of revealling the development trend of China's chemical fiber machinery,providing reference for chemical fiber machinery industry to determine the development direction.[Method/Process] The development present situation of chemical fiber machine was presented through technical origin countries and regions,distribution of major patent holders,technology focus and technical relevance.The Chemical fiber machinery hot spot evolution path,patent application trend and technology development trend were described by bibliometric methods.[Result/Conclusion] It was found that the chemical fiber machinery technology was in a transition from maturity to recession.The technical hot spot words focused on fiber,spinning,cellulose,electrospinning,nanotechnology,and static electricity,etc.Digitalization,intelligentization,and environmental protection were the key research and development directions of chemical fiber machine in the future.

[Purpose/Significance] Top talents are the core elements of modern national and urban competition.From the perspective of intelligence,this paper constructed an evaluation system for top talents and applied it in the field of superconducting materials,which provided a scientific,objective and fair evaluation method for top talents selection under the current background of mass entrepreneurship and innovation.[Method/Process] Based on the research perspective of intelligence science,using the methods of big data collecting,document retrieval,intelligence analysis and comprehensive evaluation,this paper constructed the evaluation index system of top talents which included 18 indicators from 3 dimensions.And then,it was applied in the field of superconducting materials.[Results/Conclusion] The results showed that the evaluation method for top talents was scientific and reliable.It could serve as a basic evaluation method for the selection and expert recommendation of top talents.With the help of expert consultation,its evaluation results could be reference as talent introduction,experts recommend of academic communication and project evaluation experts.

[Purpose/Significance] This paper analyzed and summarized the research progress of fine-grained opinion mining for online product reviews.Based on clarifying the main tasks,this paper discussed the key technologies,research status and development trends of fine-grained opinion mining,so as to provide suggestions for future research in this field.[Method/Process] This paper mainly adopted the method of literature review.It analyzed and discussed the relevant research work at home and abroad,extending from coarse-grained opinion mining to fine-grained opinion mining.On the basis of defining the main task of fine-grained opinion mining,this paper focused on its key technologies and current studies.[Result/Conclusion] This paper defined the main tasks of fine-grained opinion mining for online product reviews,including subjective and objective sentence classification,evaluation elements extraction and emotional polarity calculation,and summarized the key technologies involved in each task.

WeChat has become a common social tool and information source for modern people.WeChat resources have attracted the attention of researchers,but they are lack of overall research.The paper used the perspective of information management,from aspects of the information acquisition,information dissemination,information sharing and knowledge innovation to comb WeChat information resources research.The results showed that the current research was more sufficient on WeChat resources in communication and sharing,but there was still a gap for the knowledge innovation.In addition,there were shortcomings in WeChat's individual functions and research levels.