A nonconvex low-rank tensor completion model for spatiotemporal traffic data imputation. (August 2020)

• Record Type: Journal Article
• Authors: Chen, Xinyu; Yang, Jinming; Sun, Lijun
• Is Part Of: Transportation research. Volume 117(2020)
• Publication Date: 2020-08
• Subjects: Spatiotemporal traffic data -- Missing data imputation -- Low-rank tensor completion -- Truncated nuclear norm (TNN) minimization -- Nonconvex optimization; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2020.102673
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: A low-rank tensor completion framework is developed for spatiotemporal traffic. We use a truncated nuclear norm (TNN) in tensor rank approximation. The TNN-based model shows superior performance on various traffic data sets. Abstract: Sparsity and missing data problems are very common in spatiotemporal traffic data collected from various sensing systems. Making accurate imputation is critical to many applications in intelligent transportation systems. In this paper, we formulate the missing data imputation problem in spatiotemporal traffic data in a low-rank tensor completion (LRTC) framework and define a novel truncated nuclear norm (TNN) on traffic tensors of location × day × time of day. In particular, we introduce an universal rate parameter to control the degree of truncation on all tensor modes in the proposed LRTC-TNN model, and this allows us to better characterize the hidden patterns in spatiotemporal traffic data. Based on the framework of the Alternating Direction Method of Multipliers (ADMM), we present an efficient algorithm to obtain the optimal solution for each variable. We conduct numerical experiments on four spatiotemporal traffic data sets, and our results show that the proposed LRTC-TNN model outperforms many state-of-the-art imputation models with missing rates/patterns. Moreover, the proposed model also outperforms other baseline models in extreme missing scenarios.