针对信息基础设施PPP项目收益及政策不稳定性等导致PPP融资交割率低问题，本课题拟开展PPP可融资性影响因素获取和评价方法研究。首先通过文献研究梳理PPP成功因素，宏观分析与微观挖掘相结合，运用QCA方法研究可融资性影响因素。其次应用大数据技术，采用SNA方法构建PPP可融资性评价指标体系，避免评价指标间的相关性及共线性。然后通过信息基础设施PPP业务建模及画像构建，解决可融资性评价大数据组织、提取和重构问题，研究可融资性评价大数据特征选择、分类及聚类算法，解决大数据评价维度灾难难题。最后构建具备可解释性且不受变量分布假设约束的大数据评价模型，揭示大数据环境下信息基础设施PPP可融资性评价机制，形成可融资性大数据评价方法体系。本课题研究是对PPP可融资性评价理论与方法的拓展和深化，也是大数据方法在项目融资领域的应用尝试，可以为公私合作双方及金融机构开展可融资性评价提供理论依据和实践指导。

We will make the research to carry out PPP financability influencing factors acquisition and it’s evaluation methods of driven in information infrastructure project by big data for resolving the problem of low delivery rate of PPP financing. Firstly, the success and/or failure factors in PPP projects will be combed through literature review and the influencing factors of PPP financing evaluation will be summarized by Qualitative Comparative Analysis(QCA, in short), through combining macroeconomic analysis & micro-mining. Secondly, we will collect the internal and external information of PPP projects for digging the associated events and topics deeply in financability by using big data mining technology such as WEB mining、WEB crawler、text mining , etc.. Using the method of Social Network Analysis(SNA, in short)and for combining the correlation analysis and the causal analysis to avoid high correlationity and collinearity between evaluation indexes to construct evaluation index system of PPP financability. And then resolve the problem of data organization、extraction and reconstruction issues of financability evaluation by information infrastructure PPP business modeling and portraits constructing. Finally, the project would make research in the feature selection, classification and clustering algorithm of big data of financability evaluation, and complete the sample learning and training to solve dimension disaster problems. And then the project would construct evaluation models which are interpretable and independent of the variable distribution hypothesis. The project will reveal the evaluation mechanism under big data environment through case studies and simulation, and then form the big data evaluation system of PPP financability. The outcomes of our research will make an application try of using big data methods in the field of project financing, and expand and deepen the theories and methods of PPP financability evaluation, which provide theoretical foundation and practical operation support for the government, social investors and financial institutions.