中文摘要：

当前的财务危机预警研究主要着眼点是基于财务指标建立模型,然而会计舞弊或者会计不作为现象的存在,导致各个财务预警模型在以往的经济危机中纷纷失效。引入了非财务指标的一些模型中,选取的指标相对片面,难以适应对各种非财务指标具有不同敏感度的公司样本。大数据技术的出现,使得获得全面而客观信息成为可能,笔者提出了以网民为企业＂传感器＂的思想,即基于互联网上的相关在线信息,通过情感分析处理,以及统计网民信息发布频次,融合后形成传感器信号,在此基础上结合财务指标,尝试建立引入大数据指标的财务风险预警模型,并对模型的预测效果进行比较分析。检验显示基于大数据的财务风险预警模型具有更好的有效性。实验结果为相关方面预测我国上市公司财务危机提供了大数据方面的理论依据。

英文摘要：

The current financial crisis warning research focuses on model which is based on financial indicators. However,in reality accounting fraud or nonfeasance causes the failure of financial early warning model in the past economic crisis. the models selecting one-sided indicators are not suitable for all kinds of sample companies. The development of big data technology makes it possible to obtain comprehensive and objective information. This paper proposes that internet users are regarded as a sensor to enterprise. Based on the relevant online information,sensor signals are formed by mixing the analysis of emotion processing and internet information release frequency statistics together,try to bring the index of big data into financial risk early warning model,and compares the prediction results of the model analysis. We found the financial risk early warning model based on big data has better effectiveness. The results provides a theoretical basis for forecasting listed company financial crisis by big data.