随着城市交通系统的复杂性日益增加，交通安全问题受到广泛关注，对交通事件进行快速自动化分析和处理需求日益迫切。通过将交通事件的文本描述转换为向量形式，采用Word2Vec的Skip-gram模型来生成高质量的词向量，并基于这些向量构建支持向量机(SVM)分类模型，对事件的严重性和类型进行自动分类。通过对常州市武进区湖塘镇2021—2023年交通事故数据分析，实现了对该地区3 a交通事故的分类与分析，相较传统的交通事件统计分析方法，本方法语义捕捉能力更强，分类效率更高，对于交通事件的全面分析，优化事件响应和预防措施具有显著优势。

With the increasing complexity of urban traffic systems, traffic safety issues have gained widespread attention, and there is an urgent need for rapid and automated analysis and processing of traffic events. Converting the textual descriptions of traffic events into vector forms, we used Skip-gram model of Word2 Vec to generate high-quality word vectors, and based on these vectors, constructed a support vector machine(SVM) classification model to automatically categorize the severity and type of events. By analyzing traffic accident data from 2021 to 2023 in Hutang Town, Wujin District, Changzhou City, Zhejiang Province, China, we classified and analyzed traffic accidents in the region over three years. Compared to traditional statistical analysis methods of traffic events, the proposed method demonstrates stronger semantic capture capabilities and higher classification efficiency, providing significant advantages for comprehensive analysis of traffic events and optimizing response and prevention measures.