近年来互联网金融得到飞速发展。互联网金融行业产生了许多新的业务，激发了用户新的特征，同时引发了金融行业服务思维模式的转变。然而，传统的行业环境和研究条件使得互联网金融企业不能有针对性地充分识别和发挥用户价值，也不能准确地考虑用户产生的风险及其对安全隐私的顾虑。幸运的是，大数据的数据条件和研究方法为互联网金融的发展带来了良好契机。在总结现有相关研究缺口和相关支撑理论的基础之上，本项目首先研究融合多源大数据的互联网金融用户精准画像问题；并在此基础上重点研究基于全渠道的互联网金融用户价值发现与多维风险评测问题；最后根据对用户特征和用户价值的分析，通过设计算法和实验探讨互联网金融的个性化服务匹配、定制、推荐、效果测评与反馈改进。本项目的研究具有重要的理论创新和实践意义，为金融服务提供商、广大用户和监管机构带来商业利益和社会价值。

In recent years, Internet finance has been improving rapidly. It creates many novel business, and leads to new user characteristics, as well as the change of thinking of finance service providers. However, under traditional industry environment and research condition, Internet finance companies failed to adequately discern and utilize user values. Yet they felt difficult to know the risks from users and the users’ concern on safety and privacy issue. Fortunately, the data condition and analytical methodology of big data bring good opportunities to the further development of Internet finance. Based on summarizing the research gap of existing studies and relevant supporting theories, and taking advantage of multisource data fusion, this project firstly studied the accurate persona problem of Internet finance users. Then we pay main attention to the Internet finance user value discovery created in all possible application channels. We also probe the problem of multi-dimensional finance user risks, and safety and privacy perception. Finally, on foundation of user characteristics and value analysis, through algorithm design and test, this project discusses the matching scheme, service design and recommendation, performance measurement, feedback mechanism, and model improvement of personalized Internet finance service. This project has both important theoretical innovation and practical values. The achievements of this project could bring about business benefits and social values to finance service providers, users as well as regulators.