摘要
着眼具有异质性特征的庞大用户个体,尝试借鉴金融行业量化投资的理念和方法,通过客运市场大数据深度挖掘提出全新的客流监测理论和方法。从旅客用户画像标签特征中归纳生成固相、籍住、频数、行迹、质层等5大类因子体系并进行有效性检验,将所有因子分项进行子集交叉相嵌生成因子类,通过建立4个量化差分评判指标,生成因子类累进贡献度曲线,筛选出具有显著波动特征规律的主力因子类,将影响客运市场变化的主要特征客群显性化,并从因子异质性与内部市场变动和外部社会经济2个维度进行关联分析,发现籍住和频数2个因子对于客流变动具有较好的解释性。该研究成果为铁路客运大数据深度挖掘理论探索和转化应用方面提供了技术支撑,为客运营销分析对象从客流向客户转变、分析质量从定性向定量转变奠定基础。
Focusing on the huge individual users with heterogeneous characteristics,this paper tried to learn from the concept and method of quantitative investment in the financial industry,and put forward a new passenger flow monitoring theory and method through the deep mining of big data in the passenger transport market.From the characteristics of passenger user portrait labels,five factor systems including solid phase,residence,frequency,travel trace and qualitative layer were summarized and tested for effectiveness.All factor sub items were cross embedded into subsets to generate factor classes.Through the establishment of four quantitative difference evaluation indicators,the progressive contribution curve of genetic sub classes was generated to select the main factor classes with significant fluctuation characteristics.The passenger groups with main characteristics affecting the variation of passenger transport market were made explicit.The results of the correlation analysis carried out from the two dimensions of factor heterogeneity,internal market change and external social economy show that the two factors of residence and frequency have a good explanation for the change of passenger flow.The research results provide technical support for the theoretical exploration,transformation and application of deep mining of railway passenger transport big data,laying a foundation for the focus of passenger transport marketing analysis shifting from passenger flow to individual passengers,and for the analysis quality shifting from the qualitative to quantitative analysis.
作者
颜颖
叶蜀君
YAN Ying;YE Shujun(School of Economics and Management,Beijing Jiaotong University,Beijing 100044,China;Passenger Transport Department,China State Railway Group Co.,Ltd.,Beijing 100844,China)
出处
《铁道学报》
EI
CAS
CSCD
北大核心
2023年第6期16-25,共10页
Journal of the China Railway Society
基金
国家社会科学基金(B22N500010)。
关键词
旅客
量化
因子
动量
监测
passenger
quantification
factor
momentum
monitor
