互联网及搜索引擎的诞生使我们获得了获取丰富信息的途径，但同时伴随着信息的泛滥化和碎片化的问题，使人们产生了高效获取有针对性和个性化信息的需求，因此，互联网问答服务平台应运而生，随着自媒体和网络媒体的发展，问答平台还逐渐融合了社交属性从而形成了社会化问答平台，其中一些平台（如 Quora，知乎等）已成为信息和知识获取的重要渠道，为用户解决着数以万计的问题。然而在内容体量巨大，鱼龙混杂的回答信息中， 回答被用户所接受和采纳的程度也千差万别。探究哪些因素会影响用户对回答的采 纳行为，对回答者有针对性地?升影响力有着重要影响，同时也能够优化回答质量的评估 模型。 本文运用大数据方法， 以国内应用最广泛的社会化问答网站——知乎为研究对象，结合双路径理论和信息采纳模型，获取并形成 2011 年-2018 年的回答内容对应回答文本特征、 回答形式特征、 回答背景特征以及用户特征等，采用机器学习的方法构建了问答社区回答采纳行为预测模型，并进一步分析了各类特征的重要性。

The birth of the Internet and search engines has given us access to a wealth of information, but it has been accompanied by the proliferation and fragmentation of information. This has led to the need for efficient and targeted information. The service platform came into being. With the development of self-media and online media, the Q&A platform has gradually integrated social attributes to form a social Q&A platform. Some of these platforms (such as Quora, Zhihu, etc.) have become important platforms for information and knowledge acquisition, solving tens of problems for users. However, in the huge volume of content and mixed information, the answers’ adoption degrees by users are very different. Investigating which factors influence the user's adoption responses can have a significant impact on the respondents' targeted improvement of influence and can also be used to optimize the assessment model for answer quality. This paper uses the big data method and uses the most widely used social Q&A site in China—Zhihu as the research object. Combining the dual-path theory and the information adoption model, using the related answers’ data obtained on Zhihu and formed the features needed, like answer text features, answer form features, answer context features, answers features and etc. Machine learning was also used to construct a question-and-answer community response adoption model.