[Purpose/Significance] The information cocoons in mobile reading environment is a serious threaten to readers' cognitive ability,which causes the information narrowing,group polarization and so on.It has become an important issue to clarify the causes of information cocoons in mobile reading environment.This paper adopts grounded theory to analyze the causes of information cocoons in mobile reading environment combined with the semi-structured interview.[Method/Process] A framework of the influencing factors was proposed,involving 72 initial concepts,23 sub-categories,8 main categories and 4 dimensions.[Results/Conclusion] Based on the above categories and interactive relationships,a framework of the causes of information cocoons in mobile reading environment was constructed,and solutions and suggestions were put forward in the end.

[Purpose/Significance] Analyzing the key influencing factors and correlations of government data opening,and exploring the optimal realization path of government data opening will help speed up the process of government data opening.[Method/Process] Firstly,through theoretical analysis and literature review,screening the influencing factors of government data opening;secondly,with the help of NetDraw network analysis software and Delphi method,identified the main influencing factors of government data opening;Thirdly,based on the DEMATEL method to analyze the relationship between the main influencing factors,identified the key factors that affect government data opening,and then drew the adaptability landscape map of the path selection of government data opening.Finally,Y city,which has not yet launched the government data open platform,was selected as an example for simulation analysis to verify the usability of this method.[Result/Conclusion] For different levels of government data opening in different stages,the method can effectively solve the optimal path suitable for its own situation,and provide specific measures and steps.

[Purpose/Significance] Taking the smart bracelet as an example,this paper explores people's needs for personal health information organization in the field of quantitative self,and explores the contradiction between the smart bracelet platform as a service provider of personal health information organization and the actual needs of users,hoping to provide reference for the user oriented health services of relevant equipment platforms.[Method/Process] Based on the 5E classic model of user experience,this paper used mobile experience sampling method,bombed questionnaire and interview,longitudinal tracking 12 smart Bracelet users' personal health information organization needs and quantitative self behavior data.Qualitative and quantitative methods were used to analyze the data.[Result/Conclusion] In the quantitative self field,users' satisfaction with the organizational results of personal health information provided by the platform was mainly affected by the use efficiency and effectiveness;influenced by push frequency,operation complexity and attraction,the organizational function of personal health information was not continuously used;from the perspective of user demand,the fixed demand generated by users was less and sudden demand was easy to appear,etc.Based on these,the optimization suggestions were put forward.

[Purpose/Significance] At present,there have been a lot of relevant studies under the topic of influencing factors of users' switching intention,but the empirical results of different studies are different.[Method/Process] This study conducted a meta-analysis on 38 domestic and foreign empirical studies related to the influencing factors of users' intention to switch in online communities after retrieval and screening,quantitatively and comprehensively examined the influence degree of different factors on users' intention to switch in online communities,as well as the moderating variables that may cause differences in the influence degree among different studies.[Result/Conclusion] The results of the meta-analysis showed that among the 15 influencing factors included in the meta-analysis,relative deprivation,network obligation and perceived value had the strongest correlation with users' switching intention,while perceived entertainment had the weakest correlation with users' transfer intention.As a moderating variable,community type affected the relationship between perceived entertainment,dissatisfaction,attraction of alternatives,network obligation,perceived usefulness,perceived value and users' switching intention.

[Purpose/Significance] In this paper,factors that affect the usage of government open data used by scientific research users are identified,and the interaction relationship of the factors are explored.This study offers new references for improving the construction effect of government open data and the level of user utilization,and strengthen the adaptability of government data open supply and demand.[Method/Process] Based on grounded theory,the original data was obtained through semi-structured interviews,and the factors were sorted out through the three stages of open coding,main axis coding and selective coding,the paper constructed a framework model of influencing factors.[Results/Conclusions] This paper expounded the influence and mechanism of the four dimensions of data utilization ability,data preparation,scientific research task and scientific research situation on the open data utilization of scientific research government,and provided a more in-depth reference for researchers to further improve the level of government open data utilization.

[Purpose/Significance] In order to accurately describe the differential characteristics of the user groups of college students in online health community,and to help the platform operator to obtain the needs of the user groups of college students from various directions,to provide more accurate service for college students.[Method/Process] Under the guidance of user profile technology,this paper constructed an index system of user group profile of online healthy community college students based on the VALS2 model.Then,after collecting data through questionnaire survey,the empirical research was carried out by using the theory of user life cycle,factor analysis,cluster analysis and discriminant analysis.Finally,Python language wordcloud2 tool was used to visualize the tag cloud that presents the user's profile.[Result/Conclusion] The user group profile of online health community can be divided into three types:conformity,demand and experience,which can provide decision-making reference and practical support for resource optimization and accurate service of online health community.

[Purpose/Significance] This research aims to introduce emotional factor into the basis of online public opinion heat prediction to optimize the research of online public opinion heat prediction for public health emergencies.[Method/Process] Based on the current research status,this paper proposed a prediction model of online opinion heat for public health emergencies based on sentiment analysis.Taking the "×××vaccine fraud" event as an example,the model simulation was carried out.After the data was preprocessed,multivariate time series analysis was used to realize the prediction of online public opinion heat.[Results/Conclusions] Model simulation results showed online public opinion heat prediction model with sentiment value added after 5 days had the best effect,its RMSE was 223.089 and MAPE was 0.150.Online public opinion heat prediction model with sentiment value was better than online public opinion heat prediction model without sentiment value.Introducing emotional factorinto online public opinion heat prediction for public health emergencies can significantly improve its prediction effect.

[Purpose/Significance] Aiming at the problems of high-dimensional sparseness and lack of contextual semantics in microblog short text data,this paper proposes a text feature expression method that combines topic model and word embedding in order to improve the effect of microblog topic clustering.[Method/Process] This paper took Sina Weibo as the data source,combined LDA's document-topic distribution features and the weighted Word2Vec's word embedding features to construct the fusion features of microblog short text,and used K-means algorithm for topic clustering.The results were compared with single feature clustering and standard LDA topic model,and the advantages and disadvantages of the topic clustering methods were evaluated according to F1 value.[Result/Conclusion] Compared with other methods,the fusion feature topic clustering model performed best,and its F1 value reaches 83.7%.Experiment showed that the fusion features can describe the semantic information of the text more comprehensively and accurately,and can more effectively characterize the microblog text.

[Purpose/Significance] This paper tries to improve the performance of sentiment analysis under imbalance data distribution by combining class imbalance learning with deep learning.[Method/Process] An adaptive imbalance data sentiment analysis method under the LSTM deep learning framework was proposed.When the degree of imbalance was low,the samples of minority class were oversampled and then LSTM was used for deep learning training.When the degree of imbalance was high,multiple groups of equalization undersampling was carried out for the samples of majority class,and then LSTM model was learned for each group of training data respectively.Finally,the final sentiment classification result was obtained through ensemble learning.[Result/Conclusion] The experimental results on the online commodity comment corpus showed that the method can improve the deep learning performance of class imbalance data and realize the adaptive sentiment analysis of imbalance data.

[Purpose/Significance] The scientific research data reusability is becoming a distinct characteristic of modern scientific practice,this paper is to analyze the influencing factors of scientific research data reusability,and provide reference for domestic scientific research data reuse practice.[Method/Process] Based on elaborating the definition and conceptual foundations of scientific research data reusability,it deeply analyzed the five dimensions of scientific research data reusability:data curator,policy,legal,economic,and technological,then built the pentagon model of scientific research data reusability.[Result/Conclusion] Through the analysis of the pentagon model of scientific research data reusability,the scientific research data reusability improvement measures were proposed from five dimensions:the data curator dimension should focus on data literacy improvement,the legal and policy dimensions should favor the open availability of scientific data,the economics should concern how the costs associated with the process of making scientific data reusable are distributed among the stakeholders,the technological dimension should render physical and semantic barriers irrelevant.

[Purpose/Significance] Improving information service level based on artificial intelligence has become an important focus for countries to achieve innovative development and enhance national core competitiveness.Finland has taken the lead in formulating and implementing AI strategic planning and information service policies in the EU.[Method/Process] Based on text analysis of the artificial intelligence planning and information ethics policy reports issued by the Finnish government,intelligent AuroraAI project and MyData program,the paper summarized Finland's macro-level promotion of intelligent information services' target positioning,motivation mechanism,key actions and support measures.[Results/Conclusions] Studies showed that Finnish intelligent information services strengthened "Everyone is benefited" and "Everyone tries his best",which had a reference significance for China.First,it is necessary to strengthen the construction of interdisciplinary and high-level AI think tanks to provide strong support for national AI development and application decisions in the field of information service.Second,it is possible to further increase support for artificial intelligence enterprises to promote information service technology innovation.Third,in terms of information service model innovation,more attention should be paid to the service reform of public institutions based on AI.Fourth,for information service content innovation,the research on how to improve AI literacy of the entire population should be promoted systematically and deeply.

[Purpose/Significance] In the big data environment,massive isomerized teaching data puts forward new requirements for teaching knowledge management in universities.As a knowledge intensive organization,the application of big data technology in teaching knowledge management in universities will help to promote the exploration of teaching knowledge and the improvement of teaching efficiency.[Method/Process] Focusing on case study and literature survey,this paper summarized the research and practice progress of big data and university teaching knowledge management at home and abroad in recent years.On this basis,it constructed a university teaching knowledge management mode based on big data and provides implementation suggestions for universities.[Result/Conclusion] Through the collection and storage of teaching big data and knowledge,mining and development,knowledge sharing,path expansion and achievement feedback,a virtuous circle of knowledge within the organization can be realized.Therefore,it is necessary for universities to strengthen infrastructure construction,increase explicit knowledge reserve,promote tacit knowledge big data,introduce autonomous learning platform,and create a good cultural atmosphere.

[Purpose/Significance] Mobile social network has a certain stimulating effect on social Q&A platform,but it does not fully show its effect.The reason is that social Q&A platform has not yet embedded the scene elements into its service mode,resulting in a certain degree of disconnection between social services and its environment.[Method/Process] From the perspective of the application of mobile social network in social Q&A platform,the scenario elements were included in the innovation strategy of service mode of social Q&A platform.From the connotation and characteristics of the scene,the innovation of service mode of social Q&A platform was realized based on the path of "Scene-Information Expectation-Context" adaptation.[Result/Conclusion] Taking the talent introduction Q&A as an example,this paper selected the method of multiple cases to carry on the empirical study.The results showed that the scenario configuration was an effective way to innovate the service mode of the social Q&A mp;a platform,and the talent introduction can realize its service mode innovation through the scenario configuration of the social Q&A platform,so as to build the knowledge service market.

[Purpose/Significance] In order to respond to public health emergencies and accelerate the development of public health emergency management,the article studied on public health emergency combined with big data governance.[Method/Process] It built an index system by analyzing the influencing factors of big data governance capability in public health emergency,and the index weight was determined based on D-S evidence theory and AHP.[Result/Conclusion] The research results showed that the most influential factor was big data monitoring and warning capabilities in the index system.Finally,based on the research results and combined with the emergency prevention and control measures of the COVID-19,suggestions were given for public health emergency management.

[Purpose/Significance] The paper explores the influencing factors of social media users' privacy setting intention and clarifies the action relationship among the influencing factors,which can provide reference for the improvement of the privacy setting function of social media operators and the privacy protection of users.[Method/Process] Based on qualitative research method of grounded theory,the paper collected the data of 23 social media users through semi-structured interviews,applies open coding,spindle coding and selective coding to analysis the factors of influencing social media users privacy settings.The action path model of social media users privacy settings intention was put forward.[Results/Conclusion] The influencing factors of users' intention to set privacy included six dimensions:personal trait,user perception,social context,environmental factor,information factor,and technical factor.Personal trait,user perception,and social context were the internal drive variables while environmental factor,information factor,and technical factor were the external drive variables.

[Purpose/Significance] In order to help the academic community to clarify the development process and application status of the five cue theories,fill in gaps in research review of cue theories,and provide insights for scholars to grasp the thinking and directions of the future research of cue theories.[Method/Process] Combing the relevant domestic and foreign literature,the research questions were divided into three aspects:offline product quality perception,online product quality perception,and online interpersonal communication and network communication,the existing problems and the future research directions were also pointed out.[Results/Conclusion] The existing researches have limitations such as "the questionnaire survey method leads to insufficient data objectiveness","the external validity of research conclusions is insufficient",and "the model structure is not perfect".The combination of objective data and subjective data,enriching the types of experimental scenes and research objects,and introducing new moderating variables are all research directions worthy of attention in the future.

[Purpose/Significance] To help researchers grasp the progress of the application of perceived value theory in information systems research,and to provide researchers with ideas and directions.[Methodology/Process] A systematic review of relevant domestic and international literature was conducted to summarize current research results,limitations and future research directions worthy of attention.[Results/Conclusions] The application of perceived value theory in IS research mainly includes three aspects:antecedents of perceived value,IS user behaviour,IS user satisfaction and user loyalty.Meanwhile,the current research has problems such as unclear connotation of perceived value,less attention to the dynamic nature of perceived value,limited sample collection methods and research objects,etc.Future research can make progress in these aspects.

[Purpose/Significance] In order to make the academic community fully understand expectancy disconfirmation theory and its application in the field of information system,and provide direction for the follow-up related research in IS discipline.[Method/Process] By combing domestic and foreign representative research literatures about expectancy disconfirmation theory in the field of information system,this paper summarized the research achievements and existing problems.[Result/Conclusion] Expectancy disconfirmation theory is based on the cognitive model of satisfaction decision.After adding and revising key structures,it has formed a relatively perfect model to explain consumer satisfaction decision.Its application in the field of information systems mainly focuses on three aspects:online service satisfaction,online purchase behavior and information technology/system usage behavior.There are some problems in existing research,such as "the subjective emotional effect of users' has been neglected","users' behavior is usually replaced with user's intention" and so on.Users' behavior of mobile information system and discontinuous behavior etc.are the future research directions worthy of attention.

[Purpose/Significance] The purpose of this study is to help the scholars fully understand the application and development of affective events theory,to extend the research scenario in the future study,and to make the best use of this theory.[Method/Process] By searching and reviewing the relevant literature,this paper summarized the results,limitations,and prospects for the future direction.[Result/Conclusion] Since its formation,affective events theory has extended the research scenarios,redefined the affective states,and gained an insight into how affective states work.It is found that the researches in information systems mainly focus on four aspects:knowledge sharing/hiding,usage of information systems,emotion conflicts in online work,and the compliance of information security policies.Meanwhile,besides the ignorance of the traits of the subjects,the existing paper pays less attention to the type,intensity,measurement,time lags,and dynamic of affective states.Future research can improve the measurement and collection of affective data,and pay more attention to the dynamic of subjects' affective and behaviors,the interaction of elements,and the application in the Chinese context.