关键词: 距离评判, 支持向量数据描述, 单值分类, 风险预警

Abstract: Aiming at the problem of the lack of high risk projects’ data，a hybrid intelligent model of project risk prediction based on distance evaluation and Support Vector Data Description（SVDD） is proposed.By estimating the capacity of each traditional indicator in evaluating project risk，the sensitive features can be selected and input into SVDD to identify different conditions of project risk automatically.The results of the demonstration show that the model is efficient to extract the sensitive features，reduce the dimension of the data，and improve the veracity and reliability of one-class classification in intelligent risk prediction significantly.

Key words: distance evaluation, support vector data description, one-class classification, risk prediction