公众出行信息服务多源数据融合挖掘技术研究

Study on multi-source data fusion mining technology for public-travel traffic information service

分析了公众出行信息特征及影响出行线路选择的因素,设计了多源交通数据融合挖掘的系统框架,并研究了其中的多源数据相关度计算、层次化子空间聚类及联合聚类挖掘等关键技术。理论分析及实验结果表明,系统对于多源交通数据的融合分析及高维数据的降维聚类具有良好的处理能力及可扩展性。

The features of public-travel traffic information and the impact factors of travel mute se- lection were analysed. A multi-source traffic data fusion mining system framework was designed. A series of key technologies, which include the correlation degree analysis of multi-source data, the hierarchical sub-spatial clustering and co-clustering mining analysis, were studied. The theoretical analysis and field test results show that the system has good processing ability and scalability for multi-source data fusion analysis and dimension reduction.