Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-08-13 Cooperative journals: 《计算机应用研究》
Abstract: Intelligent Internet of Vehicle (IIOV) gathers relative traffic information by all kinds of on-ground sensors. The gathered data often include irregular spatial and temporal resolution, so losing data is a common problem of IIOV. In order to solve this problem, this paper proposed a kind of new approach of losing data evaluation for IIOV which was named tensor low-rank approximation(VBPCA) based on the extracting the common traffic pattern and comparing the function estimation & tensor decomposition. The approach can get the traffic patterns under the cases of losing data and the expression of low-rank. In the experiments to test the approach, it select about 1000 road segments to do the analysis. The results show that this approach has good performance on evaluation accuracy, the bias of the data set, so it is very useful for the application of intelligent internet of vehicle.