Abstract: [Purpose/Significance] By constructing the user portraits of institution accounts in the social Q&A community, this paper aims to reveal the characteristics of content production and community interaction among different groups of institution accounts, which is of significance for optimizing the content operation of institution accounts. [Design/Methodology] This paper constructs an information behavior model for institution accounts based on the framework of activity theory. By extracting and quantifying feature tags, a conceptual model of institution accounts' user profile is described, including basic information, content production, community interaction, platform functions, and topic-focused features. User clustering is achieved by combining activity tags and content quality tags, and the portrait features of each cluster are analyzed. [Findings/Conclusion] This paper constructs a user profile of institution accounts from the perspective of activity theory, and identifies four types of institution accounts based on content quality and activity. The study finds that the core elements of institution accounts content operation are the continuous and stable output of high-quality content in areas of expertise; high-quality answers can attract high-impact users, and community exposure can be improved through community management; the content operation of institution accounts focuses on a small number of thematic strategies, but when input and output are not positively related, user diversion can be achieved by properly combining other more communicative themes. [Originality/Value] This study constructs the user profile of institution accounts based on activity theory, and identifies the characteristics of content production and community interaction, which helps optimize the content operation of institution accounts and enhances the influence of the community.

Key words: User portrait, Institution accounts, Social Q&A community, Activity theory