［Purpose/Significance］To fully grasp and understand users' interests preferences,and to accurately recommend microblogs of similar interests to users,is the key to improve users' dependence on the microblog platform.In order to solve the problems of low quality of the current microblog recommendation methods,such as user interest drift and insufficient use of trust relationship among users,a personalized recommendation method based on topic and multiple trust relationship is proposed.［Methods/Process］First of all,this paper used HDP subject model to mine the subjects of the target users and their concerned users,and obtained the interests preferences of the target users and their concerned users.Secondly,by calculating the similarity and multiple trust relationship strength of the target users and their concerned users,the interests preferences of the target users was obtained.Thirdly,the target users' individual interest and group interest were weighted linearly to get the target users' comprehensive interest.Finally,according to thedistribution probabilityof the newly-published topics and the comprehensive interest degree of the target users,the interest degree of the target users was calculated and ranked in descending order,and the first Top-N recommended results were obtained.［Results/Conclusions］Experimental results showed that the method was superior to the traditional one,fully utilized the trust relationship between users,could effectively solve the problem of user interest drift,and improve the accuracy and quality of recommendation.

［Objective/Meaning］Aiming at the current academic virtual community's insufficient attention to changes in core user interest,this paper builds a user interest transfer model based on the Ebbinghaus forgetting curve to accurately describe the core user's interest transfer and improve the accuracy of personalized recommendations for core users.［Method/Process］The paper took the "emuch BBS" as the research object,constructed user influence evaluation indicators and question-and-answer network,and then respectively used entropy method and PageRank algorithm to identify core users.On this basis,the core user interest transfer model was constructed based on the Ebbinghaus Forgetting Curve,and the applicability of this model was verified.［Results/Conclusions］The core user interest transfer model constructed in this paper could better reflect the core user's interest transfer.

［Purpose/Significance］In the context of public health emergencies,the research on user standpoint identification in rumor spreading can open up a new perspective for rumor authenticity detection,and provide new ideas for rumor governance and public opinion guidance.［Method/Process］This paper took the false rumors on the microblog during the COVID-19 pandemic as the research object,aimed at improving the accuracy of the standpoint classification of comments under rumors,constructed the standardized rumor standpoint detection and analysis modeling framework.Based on the theory of Systemic Functional Linguistics(SFL),the comprehensive multidimensional feature extraction of user comments was realized and the optimal features were screened out by statistical and visual analysis.The experimental results of different model combinations were compared and analyzed and the sample factors influencing the accuracy of classification were analyzed.［Result/Conclusion］The feature combination based on ideational meta function,textual meta function and interpersonal meta function effectively combed and complemented the previous features;the optimal features rather than all features being used as the input of tree structure classifier and adopting the ensemble learning method could ensure the overall classification effect and shorten the training time;and the imbalance of data distribution significantly affected the accuracy of the comment recognition of different standpoints.The result of "other" standpoint recognition was better,and feature selection was powerful,which maked up for the disadvantage of classification of "standpoint" position data.

［Purpose/Significance］The existing clustering research of new media events focuses on the single-dimensional attributes of events,and does not consider the network structure characteristics and text distribution characteristics of event propagation.［Method/Process］This research was inspired by the concept of information entropy,and proposed an event clustering model based on network structure entropy and content distribution entropy.The model completed cross-content event similarity comparison on the basis of characterizing event network structure characteristics and content distribution characteristics,then Network Representation Learning algorithm and k-means clustering algorithm cluster the events.This paper selected 113 microblog events as the experimental objects,and used the basic attributes of the events(likes,comments,reposts,etc.)as the cluster control experimental group.［Results/Conclusion］The analysis of the experimental results showed that the model proposed in this study could capture the deeper communication and distribution characteristics of new media events.At the same time,it could improve and supplement existing similarity calculation indicators.［Originality/Value］This research can not only extract the propagation characteristics of the event from multi-dimensional levels,that is,the characteristics of the event network structure and the distribution of event content.Also it can provide support for public opinion prediction and control.The model can also monitor the communication commonality between different events through the entropy dimension of information changes to assist subsequent reflection on public opinion events.

［Purpose/Significance］In the era of knowledge economy,cross-domain cooperation among scholars has become a common phenomenon.However,knowledge demanders often run into a wall when collecting information about scholars in related fields.Therefore,it is urgent to build a perfect intellectual catalog system for scholars to promote cooperation among knowledge subjects.［Method/Process］Analyzing the advantages and disadvantages of existing scholars' information retrieval tools and its enlightenment to the construction of scholars' catalogue,this paper put forward some suggestions for the construction of intelligent catalog system for scholars.［Results/Conclusion］It is necessary and feasible to construct an intelligent directory system for scholars,which is conducive to the cross-domain cooperation among knowledge subjects and to achieve a win-win situation.

［Purpose/Significance］Event ontology is a semantic framework for describing and organizing event contents.Summarizing and analyzing the existing event ontology models can provide references for related research.［Method/Process］This paper summarized seven general event semantic models(Event,SEM,LODE,ABC,LODE,BESTDM,Event semantic model provided by Liu Zongtian),and analyzed them in detail from their characteristics and application fields.［Result/Conclusion］The development trend of event semantic models was prospected, including the development of new semantic representation dimensions and modeling perspectives,the study of the combination of horizontal continuity and vertical hierarchization of event semantic models,the study of semantic organization of the dynamic characteristics of event semantic models.

［Purpose/Significance］In view of the emerging field of Mobile Visual Search(MVS),if it directly refers to the existing scales in the field of information systems,it would face the problems of differences in application fields,usage scenarios,and scale semantics.In order to effectively measure the factors influencing on MVS user experience,a robust and practical scale for the factors influencing on MVS user experience is designed.［Method/Process］Referring to Churchill's paradigm,Limayem and other's scale development process,this paper developed the scale through initial scale formulation and evaluation,exploratory factor analysis and confirmatory factor analysis on the basis of previous studies on factors influencing on MVS user experience.［Result/Conclusion］Finally,a scale for factors influencing on MVS user experience was developed.This scale consists of 19 measurement items from five dimensions of accuracy,completeness,quickness,flexibility and empathy.Empirical analysis showed that the scale had good reliability and validity,and could provide tool reference and theoretical guidance for the later exploration of factors influencing on MVS user behavior.

［Purpose/Significance］The outbreak of the COVID-19,as a major public health emergency in the recent past,has sparked intense public discussion on the social question and answer platform.［Method/Process］This study used Python to obtain text content related to the topic of COVID-19 on the platform,using multi-dimensional scale analysis,content analysis and sentiment analysis to reveal the epidemic theme content of public concern and analyze the shift of public attention in different periods and emotional polarity in the answer text.［Results/Conclusions］The results showed that the topics of public concern were concentrated in 6 categories in two dimensions.The public's concerns were mainly concentrated in two aspects: "dynamic attention to epidemic situation" and "prevention control & epidemic transmission",of which the focus of the first stage was mainly on domestic areas,while the focus of the second phase was shifted to foreign areas.The polarity of sentiment in the answers of the two phases was mainly neutral,and the proportion of positive emotions was the lowest.The changes in the second phase compared to the first phase indicated that the overall public sentiment was positive.Based on the results of the research,we provided references and suggestions to the relevant departments for the Internet public response and social mood changes,thereby enriching the research on the Internet public response in the context of public health emergencies.

［Purpose/Significance］Obesity caused by lifestyle and diet has already become a health problem of public concern,and weight loss information plays an important role in needs of information on public fitness.This research focuses on the thematic characteristics of weight loss information needs of users in the online Q&A community,mainly focuses on the gender differences of individual in information needs on weight loss,as support the related social services.［Method/Process］The research implemented a web crawler to extract pieces of original information in the ‘Weight Loss Section’in ‘zhihu.com’,from which the LDA model was utilized to analysis.With the illustrated pipeline,demands for losing weight were differentiated from the perspective of gender differences by thematic characteristic analysis.［Result/Conclusion］The representations of the information needs on weight loss were diversified and specialized,and people who engaged in discussion and interaction of messages like this,were at the same time,seeking for emotional support and release.In the view of gender gaps,male prefers more scientific and professional information,keeping their body shape and balanced diet.Female was in favor of the content that leads a rapid change on their figure.

［Purpose/Significance］Social commerce is an emerging mode that links social functions to promote business activities.Exploring the factors that affect user information disclosure can help the stable development of enterprises.This paper explores the relationship between personality characteristics and the willingness to disclose information by taking privacy concern anxiety and privacy apathy as the mediators.［Method/Process］Sampling from users of social commerce,this study collected data through internet questionnaire,and used partial least square method to verify the research model.［Result/Conclusion］Privacy concern anxiety and privacy apathy were the strong influencing factors of information disclosure intention;Users with different personality characteristics had different emotional tendencies,and the mediating effects of emotional response between personality characteristics and disclosure intention have been confirmed.It extended the research perspective of users' willingness to disclose information,enriched the research on the internal mechanism between personality characteristics and willingness to disclose information,and systematically analyzed the mediating effect of emotional response.

［Purpose/Significance］Scientific and Technological Report is an important product of scientific research,and it is one of the strategic resources of national scientific and technology innovation.This paper aims to promote the deepening process of Scientific and Technological Report service and improve the ability of Scientific and Technological Report service,and explore the factors affecting the advancement of Scientific and Technological Report service.［Method/Process］Through grounded theoretical analysis of the in-depth interview data of 21 staffs in Scientific and Technological Report service agencies,this paper sorted out and summarized 101 concepts and 20 categories that affected the deepening of Scientific and Technological Report service,and formed 8 main categories,summed up to 4 dimensions,and finally a GPQV theoretical model of the influencing factors of Scientific and Technological Report service was built.［Results/Conclusion］The research results showed that in an innovation-driven environment,the deepening of Scientific and Technological Report service was comprehensively influenced by four dimensions,i.e.service governance,service platform,service quality and service value perception.

［Purpose/Significance］From the perspectives of patent cooperation and technology transfer,empirical research on open innovation in universities can reveal the law of open innovation and provide a reference for solving the dilemma of "high output and low transfer rate" in universities.［Method/Process］Based on the data of patent grant,licensing and transfer collected from the incoPat database,the open innovation landscape of "double first-class" universities has been displayed under the two-dimensional framework of patent cooperation and technology transfer.Spearman correlation analysis and Granger causality test were conducted on the relationship between patent cooperation and technology transfer.［Result/Conclusion］The amount of patent cooperation,transfer and grant of universities have increased in the past two decades,but the rate of open innovation has not increased but decreased.The overall open innovation of universities is at a low level,with obvious inter-school differences.The positive correlation and causal relationship between patent cooperation and technology transfer have been confirmed,indicating a synergistic effect in the process of technological innovation.

［Purpose/Significance］Non-personalized Book Recommendation Algorithm is necessary for OPAC book retrieval system,because users are always in the state of logging out.It is impossible to access user information,without which personalized recommendation algorithm cannot work efficiently.［Method/Process］Firstly,book recommendation algorithm based on semantic similarity was designed with vector space model recommending books similar to book browsing.Secondly,algorithm based on relation of same readers was introduced,which recommended books borrowed by readers who also borrowed the book that the user was browsing.Lastly,methods to merge two algorithm were discussed.［Results/Conclusion］With ten recommending books,the result of experiment on Fudan University library's book borrowed datasets showed that success rate of algorithm was 20%,such that one in five readers could find at least one books that he would borrow in recommending book list.

［Purpose/Significance］This paper explored the key areas and issues of big data industry policies made by local governments in China through a quantitative analysis of policy text.The analysis framework proposed in the paper can contribute to the understanding of features and weaknesses of big data industry policies and be useful for policy improvement.［Methods/Process］The paper adopted the method of quantitative analysis of policy texts and there were 198 big data industry policies issued by local governments in China used for data analysis.In this paper,the analysis framework consisted of three dimensions:policy function,policy subject,and policy tool.［Results/Conclusions］The results suggested that the top-level design and goal orientation of big data industry policies should be improved;policies for research and innovation of big data technologies should be strengthened;and regulations on data security governance in big data industrial should be further improved.

［Purpose/Significance］The openness of public information resources has become an inevitable choice and practical need of the government in the era of big data.Research on the policy of opening public information resources has theoretical and practical significance for the government to steadily promote the openness of public information resources.［Method/Process］The paper divided the open life cycle of public information resources into five stages,then,constructed an analysis model of my country's public information resources open policy model from the two dimensions of policy richness and satisfaction.［Result/Conclusion］The study found that different periods of the open process of public information resources showed different policy models.The period of government Internet access and information sharing met the requirements of the open process and the policy measures are rich;the period of government information disclosure was difficult to meet the requirements of the open process and the policy measures were single;the period of public information resources met the requirements of the open process,but the policy measures were not enough.

［Purpose/Signficance］Data elements have become the "industrial blood" of the digital economy.Although the overall level of government data openness in our country has improved significantly,it cannot be ignored that there are large differences in the effectiveness of construction in different regions,and its influencing factors are worth exploring in depth.［Method/Process］Using the "ecosystem theory framework",a regression analysis of 148 prefecture-city samples from 2017 to 2019 was conducted from the four dimensions of data openers,developers,consumers and the environment.［Result/Conclusion］The study found that the four variables of "data developers' development capacity","is it located in the national big data comprehensive test area","provincial coordination of open government data" and "GDP" had significant positive effects on the effectiveness of data opening construction in various local governments.Although the research had deficiencies in cross-validation and other aspects,the research results could still provide some reference for the future data opening construction of the Chinese government.

［Purpose/Significance］The city profile based on tags semantic mining is a structural view of urban multidimensional features from the perspective of public cognition,the study of its calculation and application can provide effective data support and evaluation criteria for interactive services in urban governance.［Method/Process］According to the change of the technical environment in which the city is located,the study of urban portraits can be divided into three sequential stages as follows:the traditional sociological analysis is the germination stage.At this time,the urban portrait study was gradually extended from the cognitive analysis at the substantive level to the cognitive analysis at the psychological level;The multidimensional analysis of socialized tags was a transitional period that provides a multi-angle analysis path for subsequent research based on tag resources;Research in the media era was an extended period,emphasizing the use of intelligent methods such as machine learning to mine public perception from large-scale tag resources.［Pesult/Conclusion］The research results at each stage would effectively expand the label utilization scenarios and theoretical methods,and provide a reference for systematically constructing urban image calculation and application paradigms.

［Purpose/Significance］The quantitative analysis of political texts is an emerging interdisciplinary research direction.From the perspective of the integration of text data and analysis methods,this paper systematically summarizes the latest developments in the quantitative research of political texts,aiming to support practical guidance for future research.［Method/Process］Three main quantitative methods were summarized in the different ways of how political texts were converted into data,then concluded their process,characteristics and research tasks.［Result/Conclusion］The quantitative research of political texts has developed rapidly in recent years,which mainly studied the expansion of data resources,application of multi-domain methods,and the tokenization method became increasingly popular.In the future,researchers can pay more attention to adopting more data sources for policy research,the construction of political texts corpus and validation of methods.