摘要
为挖掘绿通车查验的业务知识,以高速公路绿通车查验记录的文本信息为数据来源,提出一种不合格绿通车致因机理建模方法。首先,运用文本挖掘提取关键词作为文本特征项,绘制词云图并建立共现矩阵;其次,采用社会网络分析方法将文本特征项划分为关键特征、重要特征、次要特征和一般特征四类;最后,通过CONCOR凝聚子群分析和层次聚类,揭示文本特征项之间的关联规则。通过陕西省实例来验证模型的有效性,结果表明:分割度取值为3时凝聚子群效果最优,327个特征项被分为8个子群和137个凝聚簇。研究结果能够有效揭示文本特征项之间的耦合关系和层级结构,揭示不合格绿通车形成原因和规律,促进绿通车查验文本信息数据的资源化。
To discuss the knowledge on the expressway toll-free vehicles of fresh agricultural products,a method of unqualified vehicles cause was proposed based on the textual records of the expressway toll-free vehicles of fresh agricultural products.Firstly,text features were extracted in term of keywords using text mining technology.Thus word cloud map and co-occurrence matrix were established.Then,the text features are divided into four categories with social network analysis method,including key features,important features,minor features and general features.Finally,CONCOR cohesive subgroup analysis and hierarchical clustering were applied to reveal the association rules among text features.This study was verified in practical application by a case study of shaanxi province.The results show that the cohesive subgroup is best with the segmentation degree value of 3.Therefore,the 327 text features are divided into 8 subclusters and 137 cohesive clusters respectively.This study provides the coupling relationship and hierarchical structure with text features.The causes and rules of unqualified toll-free vehicles on fresh agricultural products are concluded by the proposed model.This study is beneficial to textual data resource on the expressway toll-free vehicles of fresh agricultural products.
作者
陈娇娜
陶伟俊
靳引利
CHEN Jiaona;TAO Weijun;JIN Yinli(School of Electronic Engineering,Xi’an Shiyou University,Xi’an 710065,China;School of Electronics and Control,Chang’an University,Xi’an 710061,China)
出处
《交通科技与经济》
2023年第6期46-53,共8页
Technology & Economy in Areas of Communications
基金
国家自然青年科学基金项目(52002315)
西安石油大学研究生创新与实践能力培养计划项目(YCS22214248)。
关键词
交通工程
绿通车
文本挖掘
社会网络分析
层次聚类
traffic engineering
toll-free vehicles of fresh agricultural products
text mining
social network analysis
hierarchical clustering
