Forecasting the traffic flows is a critical issue for researchers and practitioners in the field of transportation. However, it is very challenging since the traffic flows usually show high nonlinearities and complex patterns. Most existing traffic flow prediction methods, lacking abilities of modeling the dynamic spatial-temporal correlations of traffic data, thus cannot yield satisfactory prediction results. In this paper, we propose a novel attention based spatial-temporal graph convolutional network (ASTGCN) model to solve traffic flow forecasting problem. ASTGCN mainly consists of three independent components to respectively model three temporal properties of traffic flows, i.e., recent, daily-periodic and weekly-periodic dependencies. More specifically, each component contains two major parts: 1) the spatial-temporal attention mechanism to effectively capture the dynamic spatialtemporal correlations in traffic data; 2) the spatial-temporal convolution which simultaneously employs graph convolutions to capture the spatial patterns and common standard convolutions to describe the temporal features. The output of the three components are weighted fused to generate the final prediction results. Experiments on two real-world datasets from the Caltrans Performance Measurement System (PeMS) demonstrate that the proposed ASTGCN model outperforms the state-of-the-art baselines.

No abstract

Customized cast orientations and parameter settings of the virtual articulator according to the patient's condyles are indispensable parts of today's digital workflows in prosthodontics. This article describes a digital technique to align the intraoral scans to a virtual articulator by using a facial scanner to locate the patient's cutaneous landmarks of the arbitrary hinge axis and the reference plane, and to customize the sagittal condylar inclination of the virtual articulator through a digital protrusive interocclusal record and a dental computer‐aided design software program. It enables individual cast orientations and virtual articulator parameter settings without conventional facebow transferring and bite registration procedures and can be easily integrated with most virtual articulator systems on the market to allow clinicians and technicians to work in a complete digital workflow and facilitate customized treatment planning and dental prosthesis fabrication.

A growing body of evidence has suggested that time, from early to late, or from past to future, was represented in a spatially oriented mental time line. However, little is known about its characteristics. The present study provided the first empirical evidence to explore the symmetry of spatial representations of past and future in the mental time line. Specifically, we compared the Spatial-Temporal Association Response Codes (STARC) effects and distance effects of past and future in four experiments. Results showed that for near past and near future, STARC effects were similar (Experiment 1). For distant past, the STARC effect was significant, but not for distant future (Experiment 2). Furthermore, the distance effect in the past was significantly stronger than in the future (Experiments 3, 4). These findings supported the idea that time points are not evenly distributed in mental time line. Spatial representations of the past and the future are asymmetric, and the spatial representation of past seems stronger than future. The logarithmic pattern of internal spatial representation of past or future is also discussed.