Research On The Influence Of Communication Structure Of Discussion Threads On Problem Solving In Social Question-and-Answer Community

Social Q&A communities have developed rapidly in the context of the Internet and become an important way for users to share knowledge and solve their questions.Q&A communities provide a platform for users to freely communicate and interact with each other,and the way users communicate also affects the problem solving.At present,academics have conducted extensive and in-depth research at the individual participant level as well as the overall community level,but threads as the basic unit for discussing questions and solving problems in Q&A communities have received less attention;second,interactive communication plays a key role in problem solving in threaded discussions,and most research on communication structure stays at the stage of qualitative analysis or exploratory research,and further quantitative research is needed from The research on communication structure is mostly at the stage of qualitative analysis or exploratory research,and further quantitative research is needed to broaden the perspective.Based on this,this study selects the level of discussion threads in social Q&A communities to investigate the impact of communication structure on problem solving in discussion threads.The main research includes using data mining methods to explore the typical communication structures that exist in discussion threads,and on this basis,using econometric and qualitative comparative analysis methods to explore the impact of communication structures on problem solving.Based on social capital theory,reciprocity theory and network communication theory,this paper first extracts typical communication structures in two typical community discussion threads in China using a frequent subgraph mining algorithm,and compares the differences between valid and invalid threads using a nonparametric test based on whether the problem is solved or not.Based on this,the specific communication structures mined are quantified and combined with the concepts of variables that can characterize the communication structures in the threads(e.g.,conversation interaction degree and participation concentration)from existing studies to construct an econometric binary logit model for regression analysis as an exploration of the linear relationship between communication structures on problem solving.Finally,this paper uses a fuzzy qualitative comparative analysis(fsQCA)approach to construct a combination of antecedent conditions for variables related to communication structure,and explores the effect of communication structure on problem solving in a nonlinear and asymmetric way under a group perspective.The results show that there is a statistically significant difference in the typical communication structure between valid and invalid threads.Effective threads had more "direct reciprocity" structures of two-way communication between questioner and respondent,questioner-led "indirect reciprocity" structures,and questioner-centered "preferred connection"The econometric regressions showed that all three quantified communication structure variables(direct reciprocity,indirect reciprocity,and priority connection)had a positive effect on problem solving,and the degree of conversational interaction could also positively affect problem solving,but the effect of participation concentration on problem solving was The hypothesis of an "inverted U-shaped" relationship between participation concentration and problem solving is not valid;in the group analysis,we obtained the core condition that dialog interaction degree and indirect reciprocity can influence problem solving,and output two paths to problem solving named according to the presence or absence of direct reciprocity.When direct reciprocity is not present,the presence of conversational interactivity,indirect reciprocity,and preferential connection can contribute to problem solving;when direct reciprocity is present,the presence of conversational interactivity,participation concentration,and indirect reciprocity can contribute to problem solving.This demonstrates that there can be asymmetric combinatorial effects on problem solving among the various communication structure variables.