【 Purpose/significance】 This paper explores the factors that affect the degree of attention to the questions in the social Q&A community. The research conclusions can provide reference for the management and operation of the Q&A community【. Method/pro⁃ cess】This paper takes MetaFilter, a social Q&A community, as the research object, and firstly analyzes the full amount of structured historical data of the community, from two dimensions: community users and question content, analyzes user behavior characteristics and question content characteristics, secondly selects multidimensional feature variables as independent variables and question re⁃ sponse frequency indicating popularity as dependent variables, and finally uses the negative binomial regression method to determine the influencing factors of question popularity by analyzing the overall validity of the model (Likelihood Ratio Test)【. Result/conclusion】

The results of the experiment showed that users who have a high degree of centrality in social networks and raise questions during non work hours are more likely to gain the attention. The shorter the text length of the question itself, the fewer topics it contains, the more readable it is, and the more polarized the sentiment expressed (especially negative sentiment), the more likely it is to receive attention. Contrary to expectations, the tag popularity of question and the visible length of the layout have no significant influence on the atten⁃ tion degree of the question.【Innovation/limitation】This study constructs dynamic social networks and tag semantic networks at the level of users and question content to further reveal the factors influencing the popularity of questions raised by users.