Topic Mining And Q&A Recommendation Research In Socialized Q&A Community In UGC Environment

With the continuous development and popularization of the Internet,human communication methods and communication channels have undergone tremendous changes.From the perspective of the changes in the information environment,the development of science and technology has led to a geometric expansion of knowledge and information.This situation has reshaped the organization and understanding of the world.From a personal point of view,the rapid advancement of science and technology,especially information technology,enhances the individual's desire to understand the world,acquire knowledge and enhance oneself.It is also the desire for knowledge that enables human beings to survive and develop.Under such historical environment and technical conditions,with the development and popularization of the Internet,human wisdom and achievements can be obtained from the Internet through simple search behavior.This way of relying on the Internet to acquire knowledge,share knowledge and disseminate knowledge has changed.Humanity continues the millennium of knowledge transfer model.With the development and evolution of this networked knowledge acquisition model,in order to obtain information more conveniently,efficiently and accurately,the Internet nodes centered on knowledge inheritance and sharing adopt two different network application methods.The wiki is a network encyclopedia,which allows any user to add or edit entries to share knowledge.The second is through online knowledge sharing represented by the online Q&A community,which is a social online question and answer that allows all users to ask and answer questions.With the in-depth development of big data technology,the online Q&A community has an increasingly important position in the field of network knowledge sharing due to its interactive,extensive and real-time characteristics.However,with the rapid growth of the online Q&A community,problems such as disorderly information expansion,data fragmentation,and backup data redundancy have constrained the deepening development of online communities,resulting in users searching,screening,and utilizing knowledge while using knowledge community services.The time cost and opportunity cost are increasing,and there are also real problems such as uneven quality of questions,difficult to guarantee the quality of answers,and difficulty in being effectively explored.These phenomena also make users' knowledge of verticalization,precision and personalization.Demand has become increasingly urgent.In view of this,this paper takes the knowledge service of the community question and answer community as the background,introduces the UGC technology and method into the topic mining and question and answer recommendation of the social question and answer community,and analyzes the socialized question and answer community from the knowledge needs of the Internet users.In the topic of topic generation,answering and recommendation,a socialized Q&A community knowledge service system based on UGC technology was proposed,and the knowledge association methods of Internet users in asking questions,answering answers and sharing knowledge were discussed.The topic mining and Q&A recommendation strategies for promoting socialized Q&A community under UGC environment are proposed.This article is divided into eight chapters.Part 1 is an introduction.It mainly includes the background meaning of the topic selection of the thesis,as well as the review of domestic and foreign research.Part 2 is related concepts and theoretical foundations.The definitions,systems,theories and characteristics of the UGC and the social Q&A community are mainly explained.Part 3 is the key technology for social topic question and answer community topic extraction and question and answer recommendation in UGC environment.It mainly explains the main techniques of the research,topic mining technology,problem recommendation technology and answer recommendation technology.Among them,the reference includes both technical models represented by LDA topics and convolutional neural networks,as well as algorithms represented by clustering evolution and data coordination.Part 4 is the model construction of the social question and answer community topic and question and answer in the UGC environment.Based on the analysis of the topic and question and answer requirements of the social question-and-answer community in the eb 2.0 environment,the dynamic model of the social question-and-answer community topic and question and answer in the UGC environment is explained through the analysis of the knowledge demand situation and the demand level.The evolutionary features,relationship representations,and design constructs are described.Part 5 is an online question-and-answer community topic mining study based on the topic model.The online question-and-answer community topic of the topic model is mainly used as the experimental object,and the experimental design and result analysis are carried out with LDA algorithm and pseudo code as the key technologies.Part 6 is a question-based recommendation study for a socialized Q&A community based on multi-source hybrid tags.The socialized question and answer community of multi-source hybrid tags was mainly used as the experimental object,and the experimental design and result analysis were carried out with the key matching technology of tag matching and multi-far mixed tag library.Part 7 is an answer recommendation method based on two-way long-term and short-term memory networks.It is mainly the bidirectional long-term and short-term memory network.The recommended method is the experimental object.The convolutional neural network and its related word vector model and correlation calculation are the key technologies.The laboratory design and result analysis are carried out.Part 8 is the conclusion and outlook.Explain the conclusions and shortcomings of this study.The research shows that technical researchers can explore their features and functions through question-and-answer clustering,content extraction,semantic mining and other technologies.At the same time,through the analysis of user information and knowledge content,establish relevant models and recommendation algorithms to establish a stable platform.