Time Series and Correlation Analysis Theories-based Urban Traffic Flow Erroneous Data Imputation Models
Based on the analysis of the characteristics of the urban traffic flow erroneous data, in combination with the synthesis of the state-of-the-art on the erroneous data imputation methods, this paper first presents that: (1) the two important elements related to the erroneous data imputation are the data sources used for correction and the correction approaches, and (2) The data sources used for correction belong to those true historical data sets, which are correlated to the erroneous data either temporally or spatially. Furthermore, corresponding to the erroneous data identification approaches, this paper proposed the time series and correlation analysis theories-based urban traffic flow erroneous data imputation models. Finally, the case studies on the field data from Beijings Expressway demonstrate that the proposed erroneous data imputation model that combines the traditional time series forecasting methods with the correlation analysis theory improves the level of accuracy substantially,
Correlation Analysis Imputation Optimization Models Traffic Flow Data Time Series
Xu Wu Yanbin Geng
School of Traffic and Transportation Beijing Jiaotong University, P. R. China, 100044 Transportation Planning and Research Institute Ministry of Communications, Beijing, P. R. China, 100
国际会议
北京
英文
726-732
2007-08-18（万方平台首次上网日期，不代表论文的发表时间）