SDCCD: Spatiotemporal Data Cleaning based on Collision Detection

In the era of big data, data resources are becoming more and more abundant, and the quality of data is getting more and more attention. Data cleaning is the process of identifying and processing dirty data in order to improve the quality of data, which is conducive to make full use of the collected data and ensure the effectiveness and accuracy of the follow-up data analysis. Spatiotemporal data is a kind of time series data, and its cleaning has been widely studied. However, existing methods based on smoothing and statistics are often only suitable for dense spatiotemporal datasets. In this paper, we propose a general spatiotemporal data cleaning method based on collision detection (SDCCD), which is suitable for both dense and sparse spatiotemporal datasets. Experiments on real spatiotemporal datasets show that SDCCD can effectively detect and process spatiotemporal collision records in spatiotemporal datasets.