基于主成分分析和BP神经网络的网约车服务质量评价
Internet private hire vehicle service quality evaluation based on principal component analysis and BP neural network

摘要 为客观、科学评价网约车服务质量，从顾客满意度角度出发，通过分析网约车服务过程，建立网约车服务质量评价指标体系；为消除指标间的相关性，利用主成分分析法提取累计贡献率超过85%的主成分作为BP神经网络模型输入；通过分析BP神经网络模型的原理构建具体BP神经网络拓扑结构；以网约车乘客满意度调研问卷为数据基础，对BP神经网络模型进行训练、仿真，并与传统BP神经网络模型及SERVQUAL模型评价结果进行对比.结果表明: 本文构建模型收敛效率高、评价误差小，能够反映网约车服务质量的水平，可以为评价网约车服务质量提供有效理论支撑.
Abstract：In order to objectively and accurately evaluate the quality service of internet private hire vehicle, the evaluation index system of service quality is established by analyzing the service process from the perspective of customer satisfaction. Then, in order to eliminate the correction among indexes, the principal components of cumulative contribution rate over 85% are selected as the input of BP neural network model by principal component analysis. Further, by analyzing the principle of BP neural network model, a specific BP neural network topology is constructed. Finally, the BP neural network model is trained and simulated based on the data of passenger satisfaction questionnaire and compared with the evaluation results of traditional BP neural network model and SERVQUAL model. The results show that the proposed model is high, and the evaluation error is small which can well reflect the service quality of internet private hire vehicle and provide a theoretical basis to evaluate the service quality of internet private hire vehicle.