【摘要】 交通流量的时空数据挖掘需要完整的数据 ,因此必须处理交通流量数据中的缺失值。文章叙述了数据的缺失方式和常用的插补方法 ,根据交通流量数据时间上的周期性和空间上的相关性 ,采用平均值方法、最大期望法和数据增量法等确定性和随机性方法插补缺失数据 ,分析了这些方法的优缺点 ,并对插补结果进行比较。提出了交通缺失值插补的研究方向。更多还原

【Abstract】 Missing values in traffic flow data should be imputed because complete data are needed for space-time data mining. First, missing mechanisms and common imputati on methods are reviewed. Next, some deterministic and stochastic methods such as average value, expectation maximization and data augmentation are used in the i mputation because these data are periodic in time and correlative in space. Then advantages and disadvantages of these methods are analyzed, and imputation resu lts of them are compared. Finally, further researches on imputation methods for missing values in traffic flow data are proposed.更多还原