关键词: 财务危机预警, 灰色关联度, 案例检索, 属性权重, 相似度

Abstract: This paper proposes a financial distress pre-warning model based on Case-based Reasoning(CBR) and gray correlation degree. The method applies the gray relational analysis into CBR for business financial distress warning. The subjective and objective weights of characteristic properties are used to compute case similarity, which reflects the real similarity match. According to the important degree of each feature attribute to case retrieval, through the weight vectors remove interference for non-critical index to judge the case. Experimental results show that the sorting results of case similarity derived by the approach meet the actual situation and improve the retrieval efficiency of similar enterprises, it satisfies requirements of financial distress pre-warning.

Key words: financial distress pre-warning, gray correlation degree, case retrieval, attribute weight, similarity degree