%0 Journal Article
%T 基于贝叶斯网络的公交满意度特征分析<br>Characteristics Analysis of Bus Satisfaction Based on Bayesian Network
%A 刘天栋
%A 陈芬
%A 赵波
%J Open Journal of Transportation Technologies
%P 328-335
%@ 2326-344X
%D 2020
%I Hans Publishing
%R 10.12677/OJTT.2020.94040
%X 公交满意度是衡量公交系统设计和管理合理性的重要指标。本文从个人属性、出行信息、公交信息化、配套设施和服务质量等五个方面出发，基于贝叶斯网络，揭示它们与公交满意度之间关系，并计算这些影响因素的重要性程度。使用SPSS Modeler对南通新城区和老城区调查数据进行分析，发现它们的社会经济属性、出行目的和公交满意度具有显著地差异，提出了针对不同城区的改善方案。最后，将贝叶斯网络与TAN、Markov等模型相对比，发现该模型对于分类数据有较好的拟合能力，对样本分类的准确率也更高。<br />
Public transport satisfaction is an important index to measure the rationality of public transport system design and management. This paper starts from personal attributes, travel information, public transport information, supporting facilities and service quality, and based on Bayesian network, reveals the relationship between them and public transport satisfaction, and calculates the importance of these influencing factors. SPSS Modeler was used to analyze the survey data of Nantong’s new and old urban areas, and it was found that there were significant differences in their socio-economic attributes, travel purpose and bus satisfaction, and the improvement schemes for different urban areas were proposed. Finally, by comparing bayesian network with TAN, Markov and other models, it is found that the model has better fitting ability for classification data and higher accuracy for sample classification.
%K 公交满意度，贝叶斯网络，SPSS Modeler<br>Public Transport Satisfaction
%K Bayesian Networks
%K SPSS Modeler
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=36726