Research On Personalized Knowledge Recommendation Of Virtual Knowledge Community Based On Multidimensional Data Fusion

In the big data environment,the number of knowledge is increasing,which provides users with reference and knowledge sources,and also brings problems such as "knowledge overload" and so on.Users gradually rely on virtual knowledge community to solve the problem of knowledge acquisition and knowledge exchange.Virtual knowledge community has high-density and high-value knowledge to help users meet their knowledge needs.Users can exchange knowledge,share knowledge and accept knowledge through virtual knowledge community platform,and participate in knowledge recommendation,knowledge feedback and other knowledge services.The service mechanism of virtual knowledge community provides targeted knowledge to users and create a good knowledge interaction atmosphere,which ultimately enhances users' sense of belonging.At the same time,virtual knowledge community also has a large number of low-quality,repetitive pieces of knowledge.The quality of knowledge content is uneven;the function of knowledge information service is relatively single;the service level is relatively shallow;and the homogenization of the platform is serious.The original knowledge service theory and methods cannot fully meet the increasing diversified needs of users.Users generate knowledge content in the community while using knowledge,and they also leave a variety of forms of behavior record data.These multidimensional data contain a large number of user behavior information and personalized demand information.How to use these data to carry out knowledge service and provide users with higher quality and more adaptive service content has become an important research direction at present.From the perspective of user demand,this paper takes the virtual knowledge community as the research object,and analyzes the user demand of virtual knowledge,the data dimension division of virtual knowledge community,the mechanism and method of personalized recommendation of virtual knowledge community.This paper also explores the evaluation method of personalized knowledge recommendation effect of virtual knowledge community,and puts forward corresponding strategies and suggestions.This study was mainly carried out from six aspects as follows:(1)This paper analyzes the characteristics of user demand,including the directness and indirectness of demand expression,the real-time of demand reception,the efficiency of demand service and the reasons for its formation,which are task driven,knowledge encounter and communication interaction,so as to divide user knowledge demand into three types: knowledge demand,emotional demand and social demand.Based on the multi-theory hypothesis,it provides theoretical and empirical basis for the division of knowledge demand dimensions of virtual knowledge community users.Based on the theory of social capital,this paper proposes three hypotheses: network social capital,cognitive social capital and structural social capital,and conducts empirical research by using the method of logical regression.The results show that these three dimensions have a significant impact on the knowledge needs of virtual knowledge community users.(2)The data of virtual knowledge community is a collection of a series of interactive elements.The core of the development and design of virtual knowledge community comes from the needs of users.Based on the social capital theory,combined with the existing research,this paper analyzes the process and characteristics of multi-dimensional data fusion from the perspective of user needs,and then divides the multi-dimensional data into dimensions,including social network dimension,emotional perception dimension and user portrait dimension.It further analyzes the relationship between multidimensional data fusion and knowledge recommendation: multidimensional data fusion is the basis of knowledge recommendation,where the quality of multidimensional data fusion is the key to determine the effect of knowledge recommendation,and knowledge recommendation is the purpose of multidimensional data fusion.(3)The target positioning of personalized knowledge recommendation in virtual knowledge community is analyzed,including the availability of knowledge recommendation service,the usefulness of knowledge recommendation content,and the adaptability of knowledge recommendation results.the motivation of personalized knowledge recommendation in virtual knowledge community is discussed,including the traction of knowledge subjects,the pull of the demand for the development of knowledge and technology innovation,the promotion of the poor knowledge potential energy,and the inevitable development of the knowledge innovation environment.The components of the personalized knowledge recommendation model of the virtual knowledge community and the specific recommendation process are analyzed,including the acquisition of knowledge requirements,multi-dimensional data fusion,knowledge generation,and knowledge recommendation and absorption stages.from the four levels of data collection layer,multi-dimensional data fusion layer,knowledge aggregation layer and application layer,a multi-dimensional data fusion virtual knowledge community personalized recommendation model is constructed and discussed,with a view to exploring through the service organization model and effectively Promote user service and truly meet user needs.(4)In view of the current hot issues in the field of personalized knowledge recommendation,this chapter aims at the diverse characteristics of the virtual knowledge community and its users.By introducing three dimensions of user portrait,social network,and context awareness,the personality of the virtual knowledge community fused in multidimensional data Based on the knowledge recommendation model,the user's knowledge needs are visualized.At the same time,the personalized knowledge recommendation algorithm for the virtual knowledge community is designed using the improved maximum cluster algorithm,and the user data of a certain virtual knowledge community is selected for instance analysis to achieve accurate Personalized knowledge recommendation.The results of case analysis show that in the process of personalized knowledge recommendation of virtual knowledge community,the accuracy of personalized knowledge recommendation results can be significantly improved by introducing the three dimensions of information and constructing a user knowledge demand model by deep fusion.(5)Based on the perspective of multi-dimensional data fusion,the evaluation index system of virtual knowledge community recommendation effect is established,which aims to provide theoretical basis and evaluation criteria for the construction of virtual knowledge community recommendation service.This paper uses fuzzy analytic hierarchy process to evaluate the index system of "virtual knowledge community recommendation effect",and uses the membership function in fuzzy mathematics as a scale system to conduct empirical research on six representative virtual knowledge communities at home and abroad.The empirical results show that the indicator system has practical application value and strong operability,and can better guide the virtual knowledge community to build recommendation services.In this paper,the evaluation index system constructed at the theoretical level provides a new research perspective for the construction of virtual knowledge community recommendation services;at the practical application level,the evaluation service of recommendation services is used to improve the quality and efficiency of recommendation services of virtual knowledge communities,so as to meet user needs and Expected goals.community recommendation service,so as to meet the needs and expectations of users.(6)Based on the social capital theory,starting from the three dimensions of user portrait perspective,social network perspective,and emotion perception perspective,this paper proposes personalized recommendations and optimization strategies and recommendations for virtual knowledge communities.From the perspective of user portraits,it satisfies users 'multi-level knowledge needs,improves users' own knowledge,and uses multidimensional data fusion technology to provide users with more targeted and valuable knowledge information.From the perspective of social networks,it promotes the knowledge community 's formation and development,make the knowledge information in the platform more systematic,structured,and integrated,strengthen the communication and activities among users of the virtual knowledge community,develop social applications based on the virtual knowledge community platform,and increase the identification of community opinion leaders and user incentives The design of the mechanism promotes the exchange of knowledge among users in the community;from the perspective of emotional perception,the accuracy of personalized knowledge recommendation is achieved,the authority and integration of knowledge is strengthened,the trust and closeness of users are strengthened,and visualization is used to enrich the knowledge form,So as to make better use of the personalized knowledge recommendation effect.In the virtual knowledge community,users can obtain knowledge to meet their own knowledge expectation through the virtual knowledge community,and also expand social relations and transfer emotions through the virtual knowledge community to improve the sense of pleasure and belonging in the service process.Based on the needs of users,the multi-dimensional data of virtual knowledge community is divided into dimensions,and personalized knowledge recommendation is provided for users,so as to enhance the participation of users in virtual knowledge community,promote the sharing,utilization and innovation of knowledge,and then improve the knowledge service ability and level of virtual knowledge community.While improving the influence and competitiveness of virtual knowledge community,it also helps the virtual knowledge community develops healthily and continuously.