关键词: 用户画像, 构建模型, 多视角数据, 社会化问答平台, 生成模型, 更新模型

Abstract: ［Purpose/Significance］This study aims to explore the framework and methods of multi-view-data-driven user profile construction for social Q&A platform. It intends to provide a more comprehensive and more accurate understanding of users for service providers, which will help them improve the quality and accuracy of their services.［Design/Methodology］According to the characteristics of user data from social Q&A platform, the corresponding relationship between user data composition and user attribute extraction has been clarified from the perspectives of data mining and ontology. Then a multi-view-data-driven model of user profile construction for social Q&A platform has been proposed.［Findings/Conclusion］The model consists of four parts, which are user data acquisition, attribute sandbox construction, user profile implementation and user profile application. And the part of user profile implementation mainly involves two key steps, user profile generation and user profile update.［Originality/Value］The correspondence between multiview user data and users has been established, and the framework and methods of user profile construction for social Q&A platform have been expounded.

Key words: User profile, Model construction, Multi-view data, Social Q&A platform, Model generation, Model update