With the continuous expansion of the application field of Case-Based Reasoning (CBR) technology, it is increasingly difficult for programmers to acquire and express professional knowledge. Therefore, this paper first gives a structured expression of professional knowledge, and combines the Case-Based Reasoning method with the scientific measurement of keyword weight (Term Frequency-Inverse Document Frequency, TF-IDF). The case organization, case retrieval and case retention techniques in CBR technology were designed. And on the basis of the initial design, the inter-class dispersion calculation is added to the feature word distribution to describe the distribution of feature attributes in different categories of cases. It provides an effective method for the establishment of case reasoning model. This method is a general algorithm and can be used in many industries and fields. Taking the case of traffic congestion in urban areas as an example, the source of the management decision data of traffic congestion in urban areas has the characteristics of diversity and heterogeneity. A structured expression method and a design of case retrieval technique are presented by using the algorithm in this paper. A decision support system V1 .0 for urban road congestion was developed, and software copyright authorization was obtained.

随着基于案例的推理（CBR）技术的应用领域的不断扩展，程序员越来越难以获得和表达专业知识。因此，本文首先给出了专业知识的结构化表述，并将基于案例的推理方法与关键字权重的科学度量（术语频率-逆文档频率，TF-IDF）相结合。设计了CBR技术中的案例组织，案例检索和案例保留技术。并且在初始设计的基础上，将类别间色散计算添加到特征词分布中，以描述不同类别案例中特征属性的分布。它为案例推理模型的建立提供了有效的方法。该方法是一种通用算法，可用于许多行业和领域。以城市交通拥挤为例，城市交通拥挤管理决策数据的来源具有多样性和异质性的特点。利用该算法提出了一种结构化的表达方法和案例检索技术的设计。开发了用于城市道路拥堵的决策支持系统V1.0，并获得了软件版权授权。利用该算法提出了一种结构化的表达方法和案例检索技术的设计。开发了用于城市道路拥堵的决策支持系统V1.0，并获得了软件版权授权。提出了一种结构化的表达方法和案例检索技术的设计方法。开发了用于城市道路拥堵的决策支持系统V1.0，并获得了软件版权授权。